2020/2012(INL) | [ParlTrack]

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 1

1. Recalls that the creation and the use of artificial intelligence (AI) in the areas of culture, education and information policy raises and will continue to raise a wide range of ethical issues; calls on the European Institutions to engage in long- term thinking about the impact of AI on our democratic debates, our societies and the very nature of human beings, in order to be able to pave the way for a technology that respects our freedom, our way of life and human rights;

Amendment 10 #

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the development, the deployment and the use of AI in education, media and the cultural and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies and must aim to promote the ultimate common public good and wellness of the society;

Amendment 11 #

Laurence FARRENG, Ilana CICUREL, Irena JOVEVA, Vlad-Marius BOTOŞ, Bernard GUETTA, Radka MAXOVÁ, Morten LØKKEGAARD, Monica SEMEDO

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, media and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas, including a clear liability regime for products resulting from AI use; underlines that these criteria must be constantly adjusted to the progress in AI technologies;

Amendment 12 #

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, media, youth and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies in order for the benefits of AI to be accessible to everyone;

Amendment 13 #

Draft opinion
Paragraph 2

2. Stresses the need to develop tailor- made criteria for the use of AI in education, media and creative sectors~~, by developing ben~~ respecting the different needs and specifications of eachm arksea to ensure for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies;

Amendment 14 #

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, research, security policy, media and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies;

Amendment 15 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA, Łukasz KOHUT

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, media and creative sectors, by de~~veloping benchmarks for~~fining principles of ethically responsible and ~~accepted~~trustworthy uses of AI technologies in these areas; underlines that these criteria must be ~~constantly adjusted~~adaptable to the progress in AI technologies;

Amendment 16 #

Draft opinion
Paragraph 2 a (new)

2 a. Notes that every child enjoys the right to public quality education at all levels; therefore, calls for the development, the deployment and the use of quality AI systems that facilitate and provide quality educational tools for all at all levels and stresses that the deployment of new AI systems into schools should not lead to make a wider digital gap in society;

Amendment 17 #

Draft opinion
Paragraph 3

3. ~~Notices that AI personalised learning systems are increasingly being deployed in schools and universities~~Recognises the enormous potential that AI and robotics can contribute towards education, in particular in easing teachers' workload and addressing the increasing shortage of educational staff; notices, however, that AI-based applications for teaching and learning are, albeit slowly, being deployed at various levels of education, which is not only changing the role of teachers in the learning process ~~to one more of facilitation~~, but the processes themselves; stresses that this gradual shift should be reflected in curricula and in methods used to deliver instructions; considers that teachers’ skills need to be upgraded beyond the basic ICT competencies in order to not only harness the full potential of AI, but to also understand its limitations and to help identify the potential ethical concerns in education;

Amendment 18 #

Draft opinion
Paragraph 3

3. Notices that AI personalised learning systems are increasingly being deployed in schools and universities, which is changing the role of teachers in the learning process to one more of facilitation; stresses that this shift should be reflected in curriculacannot take the place of the educational relationship involving teachers, as assistance to pupils and students presupposes a human connection and experience of teaching methods which cannot be provided by any algorithm; considers furthermore that, in the context of the development of the digital environment, it is more essential than ever to pass on a general culture, to guarantee the freedom of people in their relations with digital tools, making it more necessary than ever for teachers to transmit it to the younger generation;

Amendment 19 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 3

3. Notices that ~~AI personalised~~technology and AI- driven learning systems are increasingly being deployed in schools and universities~~, which is changing the role of teachers in the learning process to one more of facilitation~~; emphasises that education should empower citizens to develop new forms of thinking, including AI literacy and algorithm awareness and the ability to evaluate the impact of AI on information, knowledge, and decision-making; stresses that this ~~shift~~ should be reflected in educational curricula;

Amendment 2 #

Draft opinion
Paragraph 1

1. Recalls that the ~~creation~~development, the deployment and the use of artificial intelligence (AI) in the areas of culture, education and information policy raises and will continue to raise a wide range of ethical issues; stresses that the Union should lead the way towards an ethical AI anchored in European values and ensuring the protection of the fundamental rights within a more democratic, fair and sustainable Europe;

Amendment 20 #

Loucas FOURLAS

Draft opinion
Paragraph 3

Amendment 21 #

Laurence FARRENG, Ilana CICUREL, Irena JOVEVA, Vlad-Marius BOTOŞ, Bernard GUETTA, Radka MAXOVÁ, Morten LØKKEGAARD, Monica SEMEDO

Draft opinion
Paragraph 3

3. Notices that AI personalised learning systems are increasingly being deployed in schools and universities, which is changing the role of teachers in the learning process ~~to one more of facilitation~~by individualising monitoring and teaching; stresses that this shift should be reflected in school curricula, as well as in teacher training; recalls that AI should always be a support and not a replacement for the education provided by teachers;

Amendment 22 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 3

Amendment 23 #

Draft opinion
Paragraph 3

3. Notices that AI personalised learning systems are increasingly being deployed in schools and universities, ~~which is changing~~ the role of teachers in the learning process ~~to one more of facilitation; stresses that~~is transitioning to facilitators and tutors; stresses that when students' curricula reflects this shift it should be ~~reflected in curricula~~anchored by human centric values which allow for human control and teachers' supervision;

Amendment 24 #

Draft opinion
Paragraph 3

3. Notices that specialised robotics and AI personalised learning systems are increasingly being deployed in schools and universities, which is changing the ~~role of teachers in the learning process to one more of facilitation; stresses that this shift should be reflected in curricula~~educational process; stresses that this shift should be reflected in curricula; stresses the need for a set of guidelines for the ethical aspects in education;

Amendment 25 #

Draft opinion
Paragraph 3

3. Notices that AI personalised learning systems are increasingly being deployed in schools and universities, which is gradually changing the role of teachers in the learning process to one more of facilitation; stresses that this shift should be assessed throughout and, if necessary, reflected in curricula;

Amendment 26 #

Draft opinion
Paragraph 3

3. Notices that AI personalised learning systems are increasingly being deployed in schools and universities, including specialised universities, which is changing the role of teachers in the learning process to one ~~more~~, inter alia, of facilitation; stresses that this shift should be reflected in curricula;

Amendment 27 #

Draft opinion
Paragraph 3 a (new)

3 a. Stresses that where machine learning is used in the selection of potential students, adequate safeguards must be implemented, including informing applicants of these procedures and their rights in this regard; notes that the relevant algorithms need to be trained on broad data sets in order to prevent the algorithms from unfairly discriminating against certain groups; is of the view that the relevant decisions taken with the help of automated processes need to be explainable, including, if necessary, to the rejected students;

Amendment 28 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 3 a (new)

3 a. Calls for an AI, robotics and related technologies strategy at Union level in order to transform and update our educational systems, prepare our educational institutions at all levels and equip teachers and pupils with skills and abilities; considers that the framework on ethics should also provide a set of ethical recommendations in order to help deal with AI, robotics and related technologies in education;

Amendment 29 #

Draft opinion
Paragraph 3 a (new)

3 a. Stresses that children need special protection measures related to the data that might be gathered by AI technologies, and recalls the need to adopt an ethical regulation to ensure an adequate protection level and privacy standards, in particular with regard to their educational path;

Amendment 3 #

Draft opinion
Paragraph 1

1. Recalls that the ~~creation and the~~ use of artificial intelligence (AI) in the areas of cultur~~e, education and information policy~~al and creative sectors (CCS), education as well as media and online platforms, similar to other sectors, not only has enormous potential, but has also raisesd and will continue to raise a wide range of ethical issues that need to be addressed; stresses, however, that governance of technology must be implemented in a way that does not disrupt innovation or curtail freedom of expression;

Amendment 30 #

Loucas FOURLAS

Draft opinion
Paragraph 3 a (new)

3a. Emphasises that teachers have a significant role to play in the educational process and must be familiar with the artificial intelligence systems they will be using for this purpose without, however, sidelining their role and physical presence;

Amendment 31 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 3 b (new)

3 b. Underlines the importance of using AI, robotics and related technologies in schools and universities thereby enabling them to adopt new, more efficient learning methods that will increase the success rates of pupils and students; underlines the importance of training teachers, pupils and students with the know-how regarding the ethical aspects of AI, robotics and related technologies;

Amendment 32 #

Draft opinion
Paragraph 3 b (new)

3 b. Notes that there is a need to clarify the concept of arts and cultural and creative works, as well as the role of humans as creators and artists, when AI technologies are involved in the creation of the works; stresses the need to clarify the framework of intellectual property rights applicable to AI-generated works;

Amendment 33 #

Draft opinion
Paragraph 3 c (new)

3 c. Stresses that open and equal access to AI, robotics and related technologies across the Union and within Member States is of utmost importance; considers that the Union support for AI, robotics and related technologies innovation and research should be widely available across the Union; stresses the importance, in this framework, of special support that should be given to tech developers and beneficiaries from disadvantaged and disabilities groups;

Amendment 34 #

Draft opinion
Paragraph 4

4. Acknowledges the great potential of AI in the areas of information and media; underlines that, if not regulated, it might have also ethically adverse effects by spreading fake news, creating information bubbles and exploiting biases incorporated into AI algorithms; recalls that a free and complete education is a necessary condition to enable citizens to take full advantage of the fundamental human right referred to in Article 19 of the Universal Declaration of Human Rights, which stipulates that 'Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontier';

Amendment 35 #

Draft opinion
Paragraph 4

4. Acknowledges the gr~~eat~~owing potential of AI in the areas of information ~~and media~~, media and online platforms, including as a powerful tool to fight disinformation; is concerned, however, about its dual use in the sector, especially the potential to manipulate public opinion online and the rise of AI-enabled computational propaganda and false information; underlines that, if ~~not~~AI is not properly understood and regulated, it might ~~have also~~also have ethically adverse effects ~~by spreading fake news,~~such as creating information bubbles and exploiting biases incorporated into AI algorithms;

Amendment 36 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 4

Amendment 37 #

Draft opinion
Paragraph 4

4. Acknowledges the great potential of AI in the areas of information and media; underlines that, if not regulated, itAI technologies might have ~~also~~ ethically adverse effects by ~~spreading fake news, creating information bubbles and exploiting biases incorporated into AI algorithm~~exploiting bias in data and algorithms that may lead to disseminating disinformation, creating information bubbles and spreading fake news;

Amendment 38 #

Draft opinion
Paragraph 4

4. Acknowledges the great potential of AI in the areas of information and media; underlines that~~, if not regulated, it might have also~~ particular attention must be given to the danger of misuse and possible ethically adverse effects by spreading disinformation, fake news, creating information bubbles and exploiting biases incorporated into AI algorithms;

Amendment 39 #

Draft opinion
Paragraph 4

4. Acknowledges the great potential of AI in the areas of information and media; underlines that, if not ~~regulated~~used responsibly, it might have also ethically adverse effects by spreading fake news, creating information bubbles and exploiting biases incorporated into AI algorithms;

Amendment 4 #

Draft opinion
Paragraph 1

1. Recalls that the creation and the use of artificial intelligence (AI) in the areas of culture, education and security and information policy raises and will continue to raise a wide range of ethical issues; stresses that, in connection with AI, the protection of human dignity must always be taken into account;

Amendment 40 #

Draft opinion
Paragraph 4

4. Acknowledges the great potential of AI in the areas of information and media; underlines that, if not regulated, it might have also ethically adverse effects by generating and spreading fake news, creating information bubbles and exploiting biases incorporated into AI algorithms;

Amendment 41 #

Draft opinion
Paragraph 4 a (new)

4 a. Considers that the use of certain types of AI, such as facial recognition and behaviour detection systems, may have a damaging effect on the role of media and journalists as watchdogs of democracy; underlines, therefore, that the use of those systems in public spaces should be prohibited;

Amendment 42 #

Draft opinion
Paragraph 5

Amendment 43 #

Draft opinion
Paragraph 5

5. Emphasises that med~~ucation should empower citizens~~ia competence is crucial in order to be able to critically assess and understand these new developments; thus, calls on the Member States to invest in media competence education so that all citizens, including the vulnerable social groups, are able, from an early age, to develop new forms of critical thinking, including ‘algorithm awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision-making; and so that everyone in Europe can benefit from the AI transformation of the Union economy;

Amendment 44 #

Draft opinion
Paragraph 5

5. Emphasises the need to raise awareness and understanding in the general public about the role and impact of AI through formal and non-formal education, including about the use of algorithms and their impact, inter alia, on jobs and privacy; considers that education should empower citizens to develop ~~new forms of~~ critical thinking and digital literacy, including ‘algorithm awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision-making;

Amendment 45 #

Draft opinion
Paragraph 5

5. Emphasises that education should empower citizens to develop new forms of critical thinking, including ‘algorithm awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision-making; recommends that the Commission promote AI-, robotics- and technology-related formats of education and continuous education;

Amendment 46 #

Draft opinion
Paragraph 5

5. Emphasises that education should empower citizens to develop new forms of critical thinking, including ‘algorithm awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision-making, and give them an understanding of the place occupied by IT systems in selecting, interpreting, storing and representing data;

Amendment 47 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 5

5. Emphasises that education should empower citizens to develop new forms of critical thinking, including ‘algorithm awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision-making; such qualities are, in any case, required for the study of the humanities and should continue to be cultivated;

Amendment 48 #

Draft opinion
Paragraph 5

5. Emphasises that education should empower citizens to develop new forms of critical thinking~~, including~~; calls for the establishment of digital literacy tools at all levels of education which include ‘algorithm and data awareness’ and the ability to reflect on the impact of AI on information, knowledge, and decision- making;

Amendment 49 #

Laurence FARRENG, Ilana CICUREL, Irena JOVEVA, Vlad-Marius BOTOŞ, Bernard GUETTA, Radka MAXOVÁ, Morten LØKKEGAARD, Monica SEMEDO

Draft opinion
Paragraph 5

5. Emphasises that education should empower citizens to develop new forms of critical thinking, including ‘algorithm awareness’, an understanding of the functioning of AI and its inherent biases, and the ability to reflect on the impact of AI on information, knowledge, and decision-making;

Amendment 5 #

1. Recalls that the ~~creation~~development and the use of artificial intelligence (AI) in the areas of culture, education and information policy raises and will continue to raise a wide range of ethical issues;

Amendment 50 #

Draft opinion
Paragraph 5 a (new)

5a. Considers that special attention must be given to upholding the rights of minors, given the particular influence of education on their future, specifically the right to privacy and access to education, ensuring equal opportunities in every case;

Amendment 51 #

Draft opinion
Paragraph 6

Amendment 52 #

Draft opinion
Paragraph 6

6. ENotes the important distinction between transparency of algorithms and transparency of the use of algorithms, emphasises the importance for transparency and accountability of algorithms used by AI-enabled systems such as, for example, media streaming companies, in order to ensure access to culturally and linguistically diverse content; stresses that such algorithms should not replace nor limit users' choice and should be designed in such a way that they do not privilege specific ~~works~~content by limiting their ‘personalised’ suggestions to the most popular ~~work~~content, but further enhance offers by enabling platforms to organize and present their content under different categories and on the basis of multiple factors;

Amendment 53 #

Laurence FARRENG, Ilana CICUREL, Irena JOVEVA, Vlad-Marius BOTOŞ, Bernard GUETTA, Radka MAXOVÁ, Morten LØKKEGAARD, Monica SEMEDO

Draft opinion
Paragraph 6

6. Emphasises the importance for transparency and accountability of algorithms used by media streaming companies, in order to ensure access to culturally and linguistically diverse content; believes that every user should be properly informed when an algorithm is used to recommend content and optimise his or her choices; stresses that such algorithms should be designed in such a way that they do not privilege specific works by limiting their ‘personalised’ suggestions to the most popular works; considers that any user should also be able to disable content recommendation by AI;

Amendment 54 #

Draft opinion
Paragraph 6

6. Emphasises the importance for transparency and accountability of algorithms used by media streaming companies, in order to ensure access to culturally and linguistically diverse content; stresses that such algorithms should be designed in such a way that they ~~do not~~reflect the cultural diversity of our societies and avoid privilegeing specific works by limiting their ‘personalised’ suggestions to the most popular works; proposes also that this transparency of algorithms could eventually lead to the promotion of more European works;

Amendment 55 #

Draft opinion
Paragraph 6

6. Emphasises the importance for transparency and accountability of algorithms used by media streaming companies, in order to ensure access to culturally and linguistically diverse content; stresses that such algorithms should be designed in such a way that they do not privilege specific works by limiting their ‘personalised’ suggestions to the most popular works; calls for full transparency on the algorithms used regarding creative sectors; recalls the importance of copyright protection and data protection in ethics;

Amendment 56 #

Draft opinion
Paragraph 6

6. Emphasises the importance for transparency and accountability of algorithms used by ~~media streaming compani~~video-on-demand and television broadcasting services, in order to ensure access to culturally and linguistically diverse content; ~~stresses that such algorithms should be designed in such a way that they do not privilege specific works by limiting their ‘personalised’ suggestions to the most popular works;~~

Amendment 57 #

Marcos ROS SEMPERE, Domènec RUIZ DEVESA, Łukasz KOHUT

Draft opinion
Paragraph 6

6. Emphasises the importance ~~for transparency and accountability of algorithms used by media streaming companies, in order to ensure~~of understanding algorithms used by video and music streaming services as well as spoken audio services and online e-book e-commerce stores, in order to ensure that the users also benefit from access to culturally and linguistically diverse content based on their interests; stresses that such algorithms should ~~be designed in such a way that they do~~ not privilege specific works by limiting their ‘personalised’ suggestions to solely the most popular works;

Amendment 58 #

Draft opinion
Paragraph 6

Amendment 59 #

Draft opinion
Paragraph 6

6. Emphasises the importance for transparency and accountability of algorithms used by media streaming companies, in order to ensure access to culturally and linguistically diverse content; stresses that such algorithms should be designed in such a way that they do not privilege specific works by limiting their ‘personalised’ suggestions to the most popular works; recalls that transparency of these algorithms is essential in order to avoid potentially negative repercussions of increased concentration of cultural data in the hands of major platforms for continuous media coverage, thereby ensuring access to genuine cultural openness and guaranteeing freedom of creation;

Amendment 6 #

Draft opinion
Paragraph 1

1. Recalls that the creation and the use of artificial intelligence (AI) in the areas of culture, education ~~and information policy~~, media and youth raises and will continue to raise a wide range of ethical issues;

Amendment 60 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA, Łukasz KOHUT

Draft opinion
Paragraph 6 a (new)

6 a. Emphasises that educational institutions should only use AI systems for education purposes that have been audited and certified as ethical, beneficial and acting consistently with human rights principles; reminds that open source software and open technologies are best suited for such purposes;

Amendment 61 #

Laurence FARRENG, Ilana CICUREL, Irena JOVEVA, Vlad-Marius BOTOŞ, Bernard GUETTA, Radka MAXOVÁ, Morten LØKKEGAARD, Monica SEMEDO

Draft opinion
Paragraph 6 a (new)

6 a. Insists that user data collected by AI, such as cultural preferences or educational performance, cannot be transmitted or used without the owner's knowledge;

Amendment 62 #

Draft opinion
Paragraph 6 a (new)

6 a. Notes that AI systems developed, deployed and used in the Union need to reflect its cultural diversity and its multilingualism;

Amendment 63 #

Draft opinion
Paragraph 7

7. Notes that sport has always embraced technological innovation; considers, nevertheless, that the use of AI technologies is increasingly raising questions of fair competition in sport~~; stresses that this area needs a regulatory framework.~~ whereby those teams with the most financial resources can acquire the best technology, thus potentially giving them an unfair advantage; stresses, in this regard, the need to make the relevant technology widely accessible; is of the view that further monitoring is needed to understand whether a regulatory framework in this area may be necessary in the future;

Amendment 64 #

Ibán GARCÍA DEL BLANCO, Marcos ROS SEMPERE, Domènec RUIZ DEVESA

Draft opinion
Paragraph 7

7. Notes that the use of AI technologies is increasingly raising questions of fair competition in sport; stresses that this area needs a regulatory framework.; calls for full transparency on the algorithms and technologies used in sports in order to ensure fair grounds for competition;

Amendment 65 #

Draft opinion
Paragraph 7

7. Notes that the use of AI technologies is spreading rapidly into sports competitions; therefore, it is increasingly raising questions of fair competition in sport; stresses that this area needs a regulatory framework. which applies ethical and human centric criteria in their development and use;

Amendment 66 #

Draft opinion
Paragraph 7

7. Notes that the use of AI technologies is increasingly raising questions of fair competition in sport; ~~stresses that~~emphasises that these developments have to be closely monitored in order to assess whether this area needs a regulatory framework.;

Amendment 67 #

Draft opinion
Paragraph 7

7. Notes that the use of AI technologies is increasingly raising questions of fair competition in sport; ~~stresses that this area needs a regulatory framework.~~

Amendment 68 #

Draft opinion
Paragraph 7 a (new)

7 a. Recommends the involvement of the civil society, universities, trade unions and employers associations in the process of drafting a framework on ethics and underlines the important added value of these stakeholders in the drafting of any regulatory framework; stresses the involvement of youth organisations in the process, knowing the impact this technologies will have on their future;

Amendment 69 #

Draft opinion
Paragraph 7 a (new)

7 a. Strongly believes that transparency levels should be as high as possible and proportionate to the severity of adverse human rights impact; stresses that the use of algorithmic systems in decision-making processes that carry high risks to human rights should be accompanied by particularly high levels of explainability of processes and outputs;

Amendment 7 #

Elżbieta KRUK, Ryszard Antoni LEGUTKO, Andrey SLABAKOV

Draft opinion
Paragraph 2

2. Stresses the need to develop clear, comprehensive and tailored criteria for the use of AI in education, media ~~and creative sect~~, CCS and online platforms, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to reflect the progress in AI technologies; so as to also responsibly help harness the full potential of AI; highlights in particular the need to address personal user data collection and privacy concerns in education, as well as liability issues in cases where automated learning processes lead to undesirable outcomes;

Amendment 70 #

Draft opinion
Paragraph 7 a (new)

7 a. Emphasises that opportunities provided by digitisation and new technologies, including artificial intelligence, should not lead to negligence in conservation of originals and to the disregard of traditional access to original heritage and traditional forms of promoting culture;

Amendment 71 #

Elżbieta KRUK, Ryszard Antoni LEGUTKO, Andrey SLABAKOV

Draft opinion
Paragraph 7 b (new)

7 b. Strongly believes that there is an urgent need to examine how time- honoured human rights frameworks and conventions, as well as the obligations that derive from those commitments, can guide actions and policies relating to digital cooperation and digital technology and how human rights can be meaningfully applied to ensure that no gaps in protection are caused by new and emerging technologies;

Amendment 72 #

Draft opinion
Paragraph 7 b (new)

7 b. Stresses, despite numerous advantages, opportunities and benefits presented by digitisation, new technologies and artificial intelligence, the importance of traditional forms of education and their associated social benefits; encourages Member States to promote, support and preserve traditional forms of education;

Amendment 73 #

Elżbieta KRUK, Andrey SLABAKOV

Draft opinion
Paragraph 7 c (new)

7 c. Emphasises the need to continue the fight against fake news and asks the Member States to take measures against the spread of "deepfakes" in audiovisual media;

Amendment 74 #

Draft opinion
Paragraph 7 c (new)

7 c. Encourages Member States to promote and support citizens’ participation in traditional cultural activities;

Amendment 75 #

Elżbieta KRUK, Andrey SLABAKOV

Draft opinion
Paragraph 7 d (new)

7 d. Stresses the importance of retraining workers in industries most affected by the automation of tasks and by AI; stresses that new education programmes should focus on developing skills and on the reskilling of workers so that they can seize job opportunities within the new jobs created by AI; encourages lifelong learning and the development of digital literacy programmes in order to help workers adapt to technological changes;

Amendment 76 #

Draft opinion
Paragraph 7 d (new)

7 d. Stresses the need to ensure an anthropocentric approach to AI technologies and to ensure that benefits are shared without discrimination and are accessible to all;

Amendment 77 #

Elżbieta KRUK

Draft opinion
Paragraph 7 e (new)

7 e. Calls on the Commission and the Member States to promote cooperation between the public and private sectors and academia in order to reinforce knowledge sharing, and to promote education and training on ethical implications, safety, and respect for fundamental rights, on the use of robotics and artificial intelligence, with a particular focus on human rights, safety and data privacy;

Amendment 78 #

Draft opinion
Paragraph 7 e (new)

7 e. Emphasises that AI technologies should clearly not reflect on any sort of profiling bias whether regarding identity, race, age, colour, gender or sexuality or disability;

Amendment 79 #

Draft opinion
Paragraph 7 f (new)

7 f. Recognises the need to ensure that the teams that design, develop, test and maintain, deploy and procure AI systems reflect the diversity of uses and of society in general, and that they are diverse in terms of gender, culture and age in order to mirror all essential elements of society and avoid bias;

Amendment 8 #

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, media and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies; recalls that, to provide these algorithms with a sound basis, it is necessary to impose the principles of conformity of a system with its specifications, transparency, good faith and equity;

Amendment 80 #

Draft opinion
Paragraph 7 g (new)

7 g. Highlights the role that educational systems can play in fostering and developing an ethical mindset by making people aware and informing them about AI and its use, as well as fostering AI literacy across society; underlines that educating the public to ensure proper skills should be viewed as a prerequisite before the widespread use of AI;

Amendment 81 #

Draft opinion
Paragraph 7 h (new)

7 h. Notes that along with AI technologies, education systems should also provide measures against technology addiction and lack of personal engagement and measures for responsible use by individual users;

Amendment 82 #

Draft opinion
Paragraph 7 i (new)

7 i. Recognises the threat that automation and AI might pose to employment and reiterates the need to maintain jobs as a priority, including in the cultural and creative sectors;

Amendment 83 #

Draft opinion
Paragraph 7 j (new)

7 j. Emphasises the need to examine thoroughly, properly regulate and efficiently ban the deployment of AI and the automation in political spaces, which include facial recognition, emotion recognition systems, Internet restrictions and controls to limit and restrict opposition views, controlling distributing access to public and social services, as well as disinformation or fake news, data collection, censorship and automated surveillance;

Amendment 84 #

Draft opinion
Paragraph 7 k (new)

7 k. Strongly believes that AI could have a detrimental impact on the rule of law, democracy and peoples' right of self- determination with respect to their rights to freely determine their political status and to hold opinions without interference, to exercise the right to freedom of expression, including the freedom to seek, to receive, and to impart information and ideas of all kinds;

Amendment 85 #

Draft opinion
Paragraph 7 l (new)

7 l. Notes that AI systems are software-based displaying intelligent behaviour based on the analysis of their environment; stresses that this analysis is based on statistical models of which errors form an inevitable part , sometimes with feedback loops that replicate, reinforce and prolong pre-existing biases, errors and assumptions; notes the need to ensure that systems and methods are in place to allow verification of the algorithm, explicability of the algorithm and access to remedies;

Amendment 86 #

Draft opinion
Paragraph 7 m (new)

7 m. Highlights the need to ensure that there are binding regulations laying down the rules for a whole spectrum of activity of AI, regulating all possible aspects and ensuring that principles of transparency , accountability and non-discrimination are preserved;

Amendment 87 #

Draft opinion
Paragraph 7 n (new)

7 n. Reiterates the 2019 Ethics Guidelines for Trustworthy AI and the seven key requirements for trustworthiness of AI which are (a) human agency and oversight (including fostering informed decision-making that is respectful of the individual ) (b) technical robustness and safety; (c) privacy and data governance; (d) transparency; (e) diversity, non- discrimination, and fairness; (f) societal and environmental well-being; and (g) accountability;

Amendment 88 #

Draft opinion
Paragraph 7 o (new)

7 o. Recognises that AI and automation will have an effect on the globalised economy which might be detrimental by entrenching existing inequalities and prompting regulatory arbitrage;

Amendment 9 #

Draft opinion
Paragraph 2

2. Stresses the need to develop criteria for the use of AI in education, media and creative sectors, by developing benchmarks for ethically responsible and accepted uses of AI technologies in these areas; underlines that these criteria must be constantly adjusted to the progress in AI technologies; in consultation with the competent ethics committees responsible for helping to lay the groundwork in line with European Union cultural values and legal framework provisions;

source: 650.405

2020/05/11 AFET 134 amendments...

Amendment 1 #

Draft opinion
Paragraph 1

1. Highlights that the security and defence policies of the European Union and its Member States are guided by the principles of the United Nations Charter, and by a common understanding of the universal values of the inviolable and inalienable rights of the human person, of freedom, of democracy, of equality and of the rule of law; highlights that all defence- related efforts within the Union framework respect these universal values whilst promoting peace, security and progress in Europe and in the world; recognises that the primary guarantor of Euro-Atlantic security is NATO and that decision making related to AI regulations must be made in close cooperation with Member States, the North Atlantic Council and likeminded partners as the U.S., U.K. Canada or Japan; Is of the opinion that the use of AI should be based on common set of ethical principles: responsible, equitable, traceable, reliable, and governable;

Amendment 10 #

Anna BONFRISCO, Marco ZANNI, Thierry MARIANI, Susanna CECCARDI, Lars Patrick BERG

Draft opinion
Paragraph 2

Amendment 100 #

Draft opinion
Paragraph 10

10. Underlines that the Union must promote understanding the military implications of AI, of robotics and of autonomy; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI-enabled military capabilities, in order to further strengthen Member States' technological sovereignty and strategic autonomy in this domain;

Amendment 101 #

Draft opinion
Paragraph 10

10. Underlines that the Union must promote a better understanding of the military implications, advantages and opportunities of AI, of robotics and of autonomy, including the potential for the European defence industry; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI- enabled military capabilities;

Amendment 102 #

Draft opinion
Paragraph 10

10. Underlines that the Union must promote understanding the military implications of AI, of robotics and of autonomy by working alongside military officials; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI-enabled military capabilities;

Amendment 103 #

Sven MIKSER, Raphaël GLUCKSMANN, Javi LÓPEZ, Nacho SÁNCHEZ AMOR, Eva KAILI, Claudiu MANDA, Giuliano PISAPIA, Attila ARA-KOVÁCS, Andreas SCHIEDER, Juozas OLEKAS, Isabel SANTOS, Tonino PICULA, Robert BIEDROŃ

Draft opinion
Paragraph 10

10. Underlines that the Union must promote understanding of the military implications of AI, of robotics and of autonomy; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI-enabled military capabilities;

Amendment 104 #

Draft opinion
Paragraph 10 a (new)

10a. Suggests that special attention should be paid to the technological advancement, development and deployment of drones in military operations; urges the Commission to create a code of conduct on using drone for military operations considering the potential damage they can inflict on critical infrastructures and civilians; stresses that the aforementioned code of conduct should be respected by all third countries which have economic and diplomatic ties with the Union;

Amendment 105 #

Draft opinion
Paragraph 10 a (new)

10a. Believes that enhanced cooperation between Member States and the Commission is necessary to guarantee coherent cross-border rules in the Union to encourage the collaboration between European industries and allow the development and deployment of AI- enabled technologies consistent with the prescribed safety and security standards, and the ethical framework governing the development and deployment of AI technology;

Amendment 106 #

Draft opinion
Paragraph 10 a (new)

10a. Recognises in the hybrid and advanced warfare context of today, the volume and velocity of information during the early phases of a crisis might be overwhelming for human analysts and that an AI system could process the information to ensure that human decision-makers are tracking the full spectrum of information within an appropriate time frame for a speedy response;

Amendment 107 #

Miriam LEXMANN, Tomáš ZDECHOVSKÝ, Michaela ŠOJDROVÁ, Isabel WISELER-LIMA, François-Xavier BELLAMY, Andrius KUBILIUS, Lukas MANDL

Draft opinion
Paragraph 10 a (new)

10a. Calls on the EU to institutionalise a dialogue between policy makers, developers and users of AI technology on the ethical and legal limits of AI and to inform the public at large about the advantages, the risks and the challenges posed by AI at a practical and ethical level;

Amendment 108 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 10 a (new)

10a. Underlines, that despite its added value, AI comes with a number of weaknesses, one of them relating to all sorts of different types of biases; emphasises that AI technologies should clearly be void of any sort of profiling, especially regarding gender;

Amendment 109 #

Draft opinion
Paragraph 10 a (new)

10a. Stresses the utmost importance of education and ethics-based training in the field of security and defence AI technologies with particular focus on ethics of semi-autonomous and autonomous operational systems based on human accountability;

Amendment 11 #

Draft opinion
Paragraph 2

Amendment 110 #

Draft opinion
Paragraph 10 a (new)

10a. Expresses concern about the risk of creating digital echo chambers in which people only receive information corresponding to their opinions; in this regard, expresses concern about the risk of increasing extremism and the fuelling of terrorism;

Amendment 111 #

Draft opinion
Paragraph 10 b (new)

10b. Stresses the need for a coordinated and cautious framework on the exporting of security and defence-related AI- enabled systems, products and technology to third countries; highlights the potential risks pertaining to the exporting of AI- enabled systems, products and technology to countries with authoritarian regimes or otherwise opaque systems of governance which may lead to the uncontrollable or unintended use of such applications against citizens and/or the Union; calls on the Commission to create an EU body or agency on AI tasked with the oversight, evaluation and coordination of Member States’ export activities of AI-enabled systems, products and technology in the field on security and defence to third countries;

Amendment 112 #

Draft opinion
Paragraph 10 b (new)

10b. Stresses that quantum computing could represent the most revolutionary change in conflict since the advent of atomic weaponry and thus urges that the advancement of quantum computing technologies be a priority for the Union and Member States; recognises that acts of aggression, including attacks on critical infrastructure, aided by quantum computing will create a conflict environment in which the time to make decisions will be compressed dramatically from days and hours to minutes and seconds, forcing Member States to develop capabilities that protect themselves and train both its decision makers and military personnel to respond effectively within such timeframes;

Amendment 113 #

Thierry MARIANI, Jérôme RIVIÈRE, Nicolas BAY

Draft opinion
Paragraph 10 b (new)

10b. Underlines the importance to invest in the development of human capital for artificial intelligence, fostering the necessary skills and education in an AI-enabled world; stresses in particular the importance to ensure the proper skills and training of ethicists in this field;

Amendment 114 #

Draft opinion
Paragraph 11

11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which puts in jeopardy the internal market and the objective to ensure trustworthy and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI-related projects that would help to better streamline Union efforts in this field;deleted

Amendment 115 #

Draft opinion
Paragraph 11

Amendment 116 #

Sven MIKSER, Raphaël GLUCKSMANN, Javi LÓPEZ, Nacho SÁNCHEZ AMOR, Eva KAILI, Giuliano PISAPIA, Attila ARA-KOVÁCS, Andreas SCHIEDER, Juozas OLEKAS, Isabel SANTOS, Tonino PICULA, Robert BIEDROŃ, Claudiu MANDA

Draft opinion
Paragraph 11

11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which ~~puts in jeopardy~~endangers the functioning of the internal market and the objective to ensure ~~trustworthy~~reliable and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI- related projects that would help to better streamline Union efforts in this field; stresses that AI-related projects should be synchronized with the wider EU civilian programmes devoted to AI; notes that in line with the European Commission’s White Paper on AI (COM2020/65final) excellence and testing centres concentrating on research and development of AI in the field of security and defence should be established with vigorous specifications underpinning the participation of and investment from private stakeholders;

Amendment 117 #

Draft opinion
Paragraph 11

11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which puts in jeopardy the internal market and the objective to ensure trustworthy and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI- related projects that would help to better streamline Union efforts in this field; stresses that this cooperation should be open to the most advanced and trustworthy partners outside the EU, such as the UK, US and Canada;

Amendment 118 #

Anna BONFRISCO, Marco ZANNI, Susanna CECCARDI, Lars Patrick BERG

Draft opinion
Paragraph 11

11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which puts in jeopardy the internal market and the objective to ensure trustworthy and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI- related projects that would help to better streamline Union efforts in this field, and promote at the same time the EU’s objective to strengthen human rights, international law, and multilateral solutions;

Amendment 119 #

Draft opinion
Paragraph 11

11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which puts in jeopardy the internal market and the objective to ensure trustworthy and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI- related projects that would help to better streamline Union efforts in this field, in order to strengthen Member States' technological sovereignty and strategic autonomy in this domain;

Amendment 12 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 2

2. ~~Stresses that a Union~~Recommends that a EU Member States' framework regulating the use of artificial intelligence (AI)-enabled systems in defence must respect all applicable legal regimes, in particular the international humanitarian law and the international human rights law, and be in compliance with Union law, principles and values; calls on the Union to assess the inherent AI-related risks with regard to the application of Union law, and foresee necessary adjustment and enforcement where needed;

Amendment 120 #

Dragoş TUDORACHE, Klemen GROŠELJ, Nicolae ŞTEFĂNUȚĂ, Nathalie LOISEAU, Hilde VAUTMANS, Petras AUŠTREVIČIUS

Draft opinion
Paragraph 11 a (new)

11a. Highlights that the European Union needs to strive for strategic resilience so that it is never again found unprepared in times of crisis, and underlines that especially in what artificial intelligence and its application to defence and security is concerned this is of crucial significance; emphasises that supply-chains for AI systems in defence and security that can lead to technological dependence should be recalibrated and such dependencies should be phased-out; calls for increased investment in European AI for defence and in the critical infrastructure that sustains it;

Amendment 121 #

Draft opinion
Paragraph 11 a (new)

11a. Emphasises that the development of AI that respects fundamental rights and supports the public interest requires the strategic pooling and sharing of data in the EU between private and public entities, as well as the strengthening of an EU AI ecosystem, which involves public, private, and civil society stakeholders; calls on the European Commission to foster dialogue among Member States, researchers, academics, civil society actors and the private sector so as to have inclusive policymaking processes when it comes to defence-related AI regulations;

Amendment 122 #

Nicolae ŞTEFĂNUȚĂ

Draft opinion
Paragraph 11 a (new)

11a. Highlights that, in the context of the widespread disinformation war, particularly driven by non-European actors, there is an increased need for the EU to continue investment in research, analysis, innovation and cross-border and cross-sector knowledge transfer in order to develop AI so that it could effectively contribute to combating fake news and disinformation, while at the same time respecting data privacy and the European legal framework;

Amendment 123 #

Anna BONFRISCO, Marco ZANNI, Susanna CECCARDI, Lars Patrick BERG

Draft opinion
Paragraph 11 a (new)

11a. Rejects the efforts within European Industrial Development Programme (EDIDP) and the envisioned Defence Fund (EDF) to promote research & development of autonomous weapons and disruptive military technologies; calls for an end to EU research in this field; demands the immediate termination of all calls with military AI implications like the call EDIDP-AI-2020 (Defence technologies supported by artificial intelligence);

Amendment 124 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS, Petras AUŠTREVIČIUS, Dragoş TUDORACHE, Nathalie LOISEAU

Draft opinion
Paragraph 11 a (new)

11a. Notes that since the bulk of AI research and development is happening in the private sector, it will be necessary to establish a closer cooperation with leading companies and enterprises in order to harness the potential of AI to the fullest, while fostering a better understanding of risks and benefits as well as ensuring maximum operational security;

Amendment 125 #

Draft opinion
Paragraph 11 a (new)

11a. Stresses that the EU promotes a global and comprehensive discussion on AI military applications and relevant regulatory aspects with NATO, with particular regard to interoperability in the European defence, in order to include AI into the common actions that form part of the Joint Declarations;

Amendment 126 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 11 a (new)

11a. Calls on the Commission to present their "Reinforcement of the Skills Agenda", announced in the White Paper on Artificial Intelligence on the 19th February2020, as soon as possible - in order to ensure that everyone in Europe can benefit from the digital transformations of the EU economy;

Amendment 127 #

Draft opinion
Paragraph 11 a (new)

11a. Calls on the European Commission to establish a Working Group on Security and Defence within the High-Level Expert Group on Artificial Intelligence that should specifically deal with policy and investment questions as well as ethical aspects of AI in the field of security and defence;

Amendment 128 #

Draft opinion
Paragraph 11 a (new)

11a. Recommends that the AI defence R&D should focus on defence systems development and discourage the creation of weaponised autonomous robots which could be characterised by mass destruction capabilities;

Amendment 129 #

Draft opinion
Paragraph 11 b (new)

11b. Stresses the importance of the creation of an ethical code of conduct underpinning the deployment of weaponised AI-enabled systems in military operations, similar to the existing regulatory framework prohibiting the deployment of chemical and biological weapons; is of the opinion that the Commission should initiate the creation of standards on the use of AI-enabled weapons systems in warfare according to international humanitarian law;

Amendment 13 #

Dragoş TUDORACHE, Klemen GROŠELJ, Nicolae ŞTEFĂNUȚĂ, Hilde VAUTMANS, Petras AUŠTREVIČIUS

Draft opinion
Paragraph 2 a (new)

2a. Recognises that unlike defence industrial bases, critical AI innovations could come from small Member States, thus a CSDP-standardized approach should ensure that smaller Member States and SME’s are not crowded out. Stresses that a set of common EU AI capabilities matched to a Member States operating concepts can bridge the technical gaps that could leave out states lacking the relevant technology, industry expertise or the ability to implement AI systems in their defence ministries;

Amendment 130 #

Draft opinion
Paragraph 11 b (new)

11b. Stresses that, as a global actor, the European Union should pursue the international adoption of its ethical and technical standards in AI-powered defence systems; considers that the Union should engage in AI diplomacy in international fora with likeminded partners like the G7, the G20,and the OECD;

Amendment 131 #

Dragoş TUDORACHE, Klemen GROŠELJ, Nicolae ŞTEFĂNUȚĂ, Hilde VAUTMANS, Petras AUŠTREVIČIUS

Draft opinion
Paragraph 11 c (new)

11c. Calls for increased cooperation within the NATO Alliance for the establishment of common standards and interoperability of AI systems in defence; stresses that the transatlantic relationship is crucial in preserving shared values and in countering future and emerging threats;

Amendment 132 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER

Draft opinion
Paragraph 12

12. CTakes note of the European Commission's White Paper on Artificial Intelligence and regrets that military aspects were not taken into account; calls on the European Commission and on the VP/HR to present, also as part of an overall approach, a sectoral AI strategy for defence-related activities within the Union framework, that should propose a consistent regulatory approach spanning from the inception of AI-enabled systems to their military uses and clarify the EU's position on lethal autonomous weapons systems (LAWS), respecting its resolution of 12 September 2018 on autonomous weapon systems (2018/2752(RSP)); calls on the Council, the European Commission and on the VP/HR to enter in a structured dialogue with the European Parliament to that end.

Amendment 133 #

Draft opinion
Paragraph 12

12. Calls on the European Commission and on the VP/HR to present, also as part of an overall approach, a sectoral AI strategy for defence-related activities within the Union framework, that ensures both respect for citizens’ rights and EU’s strategic interests that should propose a consistent regulatory approach spanning from the inception of AI-enabled systems to their military uses; calls on these regulatory efforts to be supported by meaningful monitoring schemes, so that normative frameworks are not outplaced by technological development and new methods of warfare; calls on the Council, the European Commission and on the VP/HR to enter in a structured dialogue with the European Parliament to that end.

Amendment 134 #

Draft opinion
Paragraph 12

12. Calls on the European Commission and on the VP/HR to present, also as part of an overall approach, a sectoral AI strategy for defence-related activities within the Union framework, that should propose a consistent ~~regulatory~~ approach spanning from the inception of AI-enabled systems to their military uses; calls on the Council, the European Commission and on the VP/HR to enter in a structured dialogue with the European Parliament to that end.

Amendment 14 #

Draft opinion
Paragraph 2 a (new)

2a. Emphasises that the geographical scope of such a framework should cover all the components of artificial intelligence, robotics and related technologies developed, deployed or used in the Union, including in cases where part of the technologies might be located outside the Union or not have a specific location;

Amendment 15 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 2 a (new)

2a. Underlines the need of constant monitoring of the use of AI; especially from the point of view of its advantages and disadvantages, as well as, its impact on the protection of Universal Human Rights;

Amendment 16 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 2 b (new)

2b. Recalls that the impact of AI systems should be considered not only from an individual perspective but also from the perspective of a society as a whole; calls to fully incorporate in a new framework, the human-centric approach based on the Communication on Building Trust in Human-Centric AI and the input obtained in the Ethics Guidelines prepared by the High-Level Expert Group on AI;

Amendment 17 #

Draft opinion
Paragraph 3

3. Considers that current and future defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume leading role in research and development of AI systems in defence field; believes that the use of AI-enabled applications in defence offer number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision- making, as well as greater reliability of military equipment; recalls that AI systems are also becoming key elements in countering emerging security threats;deleted

Amendment 18 #

Draft opinion
Paragraph 3

3. Considers that ~~current and future defence-related activities within~~AI, deep learning, quantum computing, machine learning, computer vision, neuro-linguistic programming, virtual reality and augmented reality are all part of the future battlespace that the Union framework ~~will~~must draw on ~~AI, on robotics and autonomy, and on related technologies and~~and prepare for; Thus considers that the Union ~~must~~could assume a leading role in research and development of AI systems in the defence field; believes that the use of AI-enabled applications in defence offer a number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision- making, as well as greater reliability of military equipment; recalls that AI systems are also becoming key elements in countering emerging security threats; encourages the European Defence Agency (EDA) to continue its developments in this space;

Amendment 19 #

Draft opinion
Paragraph 3

3. Considers that current and future defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume a leading role in research and development of AI systems in the defence field in order to achieve technological sovereignty and strategic autonomy; believes that the use of AI-enabled applications in defence could offer a number of direct benefits to the commander such as higher quality collected data, greater situational awareness, increased speed for decision- making, as well as greater reliability of military equipment and hence reduced human risk and human casualties; recalls that AI systems are also becoming key elements in countering emerging security threats such as cyber and hybrid warfare;

Amendment 2 #

Sven MIKSER, Javi LÓPEZ, Raphaël GLUCKSMANN, Nacho SÁNCHEZ AMOR, Eva KAILI, Claudiu MANDA, Giuliano PISAPIA, Attila ARA-KOVÁCS, Andreas SCHIEDER, Juozas OLEKAS, Isabel SANTOS, Tonino PICULA

Draft opinion
Paragraph 1

1. Highlights that the security and defence policies of the European Union and its Member States are guided by the principles enshrined in the European Charter of Fundamental Rights and those of the United Nations Charter, and by a common understanding of the universal values of the inviolable and inalienable rights of the human person, of freedom, of democracy, of equality and of the rule of law; highlights that all defence- related efforts within the Union framework must respect these universal values whilst promoting peace, security and progress in Europe and in the world;

Amendment 20 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS

Draft opinion
Paragraph 3

3. Considers that current and future defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume leading role in research and development of AI systems in defence field; believes that the use of AI-enabled applications in defence offer number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision-making, as well as greater reliability of military equipment; recalls that AI systems are also becoming key elements in countering emerging security threats both in the online and offline spheres; notes, however, that AI could be exposed to forms of dangerous malign manipulation in unpredictable ways and with incalculable consequences;

Amendment 21 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 3

3. Considers that current and future defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume leading role in research and development of AI systems in defence field; believes that the use of AI-enabled applications in defence offer number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision-making, as well as greater reliability of military equipment; stresses that the development of reliable AI in the field of defence is essential for ensuring European strategic autonomy in capability and operational areas; recalls that AI systems are also becoming key elements in countering emerging security threats;

Amendment 22 #

Draft opinion
Paragraph 3

3. Considers that current and future defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume leading role in research and development of AI systems in defence field; believes that the use of AI-enabled applications in defence offer number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision-making, reduced risk of collateral damage thanks to better cabling, protection of forces on the ground, as well as greater reliability of military equipment; recalls that AI systems are also becoming key elements in countering emerging security threats;

Amendment 23 #

Miriam LEXMANN, Tomáš ZDECHOVSKÝ, Michaela ŠOJDROVÁ, Isabel WISELER-LIMA, François-Xavier BELLAMY, Andrius KUBILIUS, Lukas MANDL

Draft opinion
Paragraph 3

3. Considers that current and future security and defence-related activities within the Union framework will draw on AI, on robotics and autonomy, and on related technologies and that the Union must assume leading role in research and development of AI systems in security and defence field; believes that the use of AI- enabled applications in security and defence offer number of direct benefits such as higher quality collected data, greater situational awareness, increased speed for decision-making, as well as greater reliability of military equipment; recalls that AI systems are also becoming key elements in countering emerging and hybrid security threats;

Amendment 24 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 3 a (new)

3a. Calls for establishing of synergies and networks between the various European research centres on AI as well as other multilateral fora, such as: Council of Europe, the United Nations Educational Scientific and Cultural Organization (UNESCO), the Organisation for Economic Co-operation and Development’s (OECD),the World Trade Organisation and the International Telecommunications Union (ITU), in order to align their efforts and to better coordinate the developments of the AI technology;

Amendment 25 #

Draft opinion
Paragraph 3 a (new)

3a. Is concerned that the research & development of disruptive military technologies and autonomous weapons play a more and more dominant role in the Unions security policy; is convinced that even partially autonomous systems still force the opponent to further automate its own systems and will lead to a further uncontrollable arms race, which inevitably leads to the introduction of killer robots at the end;

Amendment 26 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 3 a (new)

3a. Underlines at the same time all the risks and challenges of unregulated use of AI; strongly believes technological advances in the field have to go hand in hand with an ample discussion on the ethical aspects of the use of AI and its impacts on societies and communities, including in the field of security and defence;

Amendment 27 #

Draft opinion
Paragraph 3 a (new)

3a. Stresses that AI technologies are, in essence, dual use; highlights that AI in defence-related activities is a transverse disruptive technology whose development may provide opportunities for the competitiveness and the strategic autonomy of the EU;

Amendment 28 #

Draft opinion
Paragraph 3 a (new)

3a. Highlights the importance of exploring and developing synergies between civilian and military AI research and applications, particularly from a total defence perspective;

Amendment 29 #

Miriam LEXMANN, Tomáš ZDECHOVSKÝ, Michaela ŠOJDROVÁ, Isabel WISELER-LIMA, François-Xavier BELLAMY, Andrius KUBILIUS, Lukas MANDL

3a. Stresses that AI technologies are of dual use: the development of AI in defence-related activities benefits from exchanges between military and civil technologies;

Amendment 3 #

Draft opinion
Paragraph 1

Amendment 30 #

Draft opinion
Paragraph 3 a (new)

3a. Is of the opinion that reliable, robust and trustworthy AI is a foundational requirement for modern and effective military of XXI century;

Amendment 31 #

Draft opinion
Paragraph 3 b (new)

3b. Calls for a total ban on all autonomous weapons systems, regardless of the degree of automation, enshrined in international law, for example in the UN Convention on Conventional Weapons (CCW) or in a newly drafted convention banning these weapons; further calls for an end to all research into the gradual automation of weapons systems;

Amendment 32 #

Draft opinion
Paragraph 3 b (new)

3b. Considers, while ethical and moral considerations must be examined while developing artificial intelligence, it is important to highlight and communicate the economic and societal benefits that could be generated by AI;

Amendment 33 #

Thierry MARIANI, Jérôme RIVIÈRE, Nicolas BAY

Draft opinion
Paragraph 3 c (new)

3c. Recalls that the European Parliament already called twice (2014 1a and 20181b) for a ban on (lethal) autonomous weapons; _________________ 1a https://www.europarl.europa.eu/sides/get Doc.do?pubRef=-//EP//TEXT+TA+P7- TA-2014-0172+0+DOC+XML+V0//EN 1b https://www.europarl.europa.eu/doceo/doc ument/TA-8-2018-0341_EN.html

Amendment 34 #

Draft opinion
Paragraph 4

Amendment 35 #

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility, of protecting our citizens, and of defending our values, whilst seizing the opportunities that those technologies offer; calls on Member States to combine the analytical efforts of CSDP, NATO and individual command structures to ascertain how, and when, a scenario will develop; such machine-learning crisis simulation systems could offer improved visibility into the causes and drivers of a crisis that might otherwise be overlooked by conventional analysis, which can be too narrow to capture the true complexity of a situation;

Amendment 36 #

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, ~~the Union~~whereby technology fully respects Human Rights and humans retain authority over automated decision- making systems, the Union needs a robust AI regulatory framework focused on security and defence, followsing a path of responsibility, of protecting our citizens, and of defending our values, that its policies aim at preserving peace, preventing conflicts and strengthening international security, whilst seizing the opportunities that those technologies offer;

Amendment 37 #

Miriam LEXMANN, Tomáš ZDECHOVSKÝ, Michaela ŠOJDROVÁ, Isabel WISELER-LIMA, François-Xavier BELLAMY, Andrius KUBILIUS, Lukas MANDL

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility, of protecting our citizens, and of defending our values, whilst seizing the opportunities that those technologies offer; welcomes and supports the Commission’s High-Level Expert Group on Artificial Intelligence ‘Ethics Guidelines for Trustworthy AI’ published on 9 April 2019 and its position on lethal autonomous weapon systems (LAWS);

Amendment 38 #

Bart GROOTHUIS, Urmas PAET, Nathalie LOISEAU, Hilde VAUTMANS, Klemen GROŠELJ

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility, of protecting our citizens, and of defending our values, whilst seizing the opportunities that those technologies offer as well as realizing that AI enabled systems will be a key element in future defence-developments and defensive capabilities;

Amendment 39 #

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility and transparency, of protecting our citizens, and of defending our values, whilst seizing the opportunities that those technologies offer;

Amendment 4 #

Anna BONFRISCO, Marco ZANNI, Thierry MARIANI, Susanna CECCARDI, Lars Patrick BERG

Draft opinion
Paragraph 1

Amendment 40 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS, Dragoş TUDORACHE

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility, of protecting our citizens and their data, and of defending our values, whilst seizing the opportunities that those technologies offer;

Amendment 41 #

Draft opinion
Paragraph 4

4. Highlights that, based on a human- centric approach, the Union follows a path of responsibility, of protecting our citizens, and of defending our values, whilst fully seizing the opportunities that those technologies offer;

Amendment 42 #

Draft opinion
Paragraph 4 a (new)

4a. Insists that a human centric approach means that humans interacting with AI systems must be able to keep full and effective self-determination over themselves, that AI systems should not unjustifiably subordinate, coerce, deceive, manipulate, condition or herd humans but they have to be designed to augment, complement, empower and support human autonomy and decision making, securing a full human oversight over work processes in AI systems at all times;

Amendment 43 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 4 a (new)

4a. Calls on the Member States and the European Commission to ensure that the algorithms used in defence systems, while keeping the necessary confidentiality, are governed by the principle of transparency, including a clear liability regime for the results of AI use; underlines that these algorithms must be constantly adjusted to the progress in AI technologies;

Amendment 44 #

Draft opinion
Paragraph 4 a (new)

4a. Stresses that not all members of the international community will follow the regulatory human-centric approach devised by the Union and that authoritarian states will devise a counter framework that will pursue, at the expense of Member States, to deliver military advantages and superiority within the AI domain;

Amendment 45 #

Draft opinion
Paragraph 4 b (new)

4b. Urges Member States to track and assess the development of AI technologies, particularly military and surveillance, within authoritarian states that avoid compliance with EU led regulations, to avoid a scenario in which our societies, militaries and institutions, fall vulnerable to hybrid warfare;

Amendment 46 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS, Petras AUŠTREVIČIUS, Dragoş TUDORACHE

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that contains the inherent risks of these technologies and prevents use for malicious purposes; those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights or physical harm; underlines that further AI research and development should ensure that AI enabled systems are better equipped to understand unique context; Calls for regular monitoring to make sure that AI- driven decision-making does not lead to discrimination and that AI systems are not trained on bias data;

Amendment 47 #

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that contains the inherent risks of these technologies and prevents use for malicious purposes; the Union working together with the Member States must determine the appropriate liability regimes applicable to innovations in AI and other immersive technologies in the field of security and defence thus establishing a legal basis for accountability and traceability mechanisms, those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights;

Amendment 48 #

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that contains the inherent risks of these technologies and prevents use for malicious purposes; those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights; urges the VP/HR, the Member States and the Council to initiate multilateral negotiations on a legally binding instrument regulating LAWS, with an effective enforcement mechanisms; regrets the failure to agree on such regulation at the Convention on Conventional Weapons;

Amendment 49 #

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes ~~for their use~~and legislation for their use while avoiding overregulation, for understanding the related ethical aspects and for fostering an ~~effective~~ international rleg~~ulatory~~al framework that ensures meaningful human control over autonomous weapon systems in order to contains the inherent risks of these technologies and prevents use for malicious purposes; ~~those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights~~calls, to that end, on the EU and its Member State to promote the discussion on lethal autonomous weapons systems (LAWS) in the UN CCW framework and other relevant fora;

Amendment 5 #

Kinga GÁL

Draft opinion
Paragraph 1 a (new)

1a. Underlines the opportunities offered by artificial intelligence, already having an impact on our world in many different ways; artificial intelligence has the potential to become the engine of productivity and economic growth; it has the ability to increase the efficiency with which things are done and improve the decision-making process by analysing large amounts of data; emphasizing in the same time the need to maintain at all times the responsible human monitoring and control in relation to artificial intelligence systems;

Amendment 50 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that consults with military, industry, law enforcement, academia and civil society stakeholders to ensure that any framework contains the inherent risks and advantages of these technologies and prevents use for malicious purposes; those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights;

Amendment 51 #

Draft opinion
Paragraph 5

5. Underlines that the Union, in connection with the work carried out by the Convention on Certain Conventional Weapons’ Governmental Expert Group, must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that contains the inherent risks of these technologies and prevents use for malicious purposes; those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights;

Amendment 52 #

Draft opinion
Paragraph 5

5. Underlines that the Union must be at the forefront ~~of mastering~~to support multilateral efforts in the UN CCW framework and other relevant fora, to discuss an effective international regulatory framework, to master those technologies by establishing well defined processes for their use, ~~for~~to understand~~ing~~ the related ethical aspects and ~~for fostering an effective international regulatory framework tha~~to contain~~s the~~ inherent risks of these technologies and prevents use for malicious purposes; those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights;

Amendment 53 #

Draft opinion
Paragraph 5 a (new)

5a. Underlines that in an era of decentralised innovation and software- driven warfare, accessible data provides strategic advantage, making AI particularly significant when combined with virtual- and augmented-reality visualisation, allowing it to play a significant role in providing advanced training and pre-deployment unit-level preparation for EU or NATO-led forces during peacetime, to ensure a rapid yet smooth transition into conducting operations; encourages therefore the development of European data, generated within the European Union’s borders, in particular with a view to developing industrial technology using machine- generated data;

Amendment 54 #

Draft opinion
Paragraph 5 a (new)

5a. Calls on the EU to work towards achieving convergence with like-minded third countries in the field pertaining the ethic of AI as well as to keep contributing in multilateral fora to build a consensus on a human-centric AI, with the goal of fostering an international regulatory framework putting a human-centric approach at its core;

Amendment 55 #

Draft opinion
Paragraph 5 a (new)

5a. Highlights that, while international discussions have largely centred on the potential development of lethal autonomous weapons systems (LAWS), ethical aspects of other application fields for AI in the defence sector such as intelligence, surveillance, and reconnaissance (ISR) or cyber operations must not be overlooked;

Amendment 56 #

Draft opinion
Paragraph 5 a (new)

5a. Endorses the key principle “ethics- by-design”, by which ethical principles are embedded into AI products and services from the outset of the design process;

Amendment 57 #

Draft opinion
Paragraph 6

6. Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible ~~trustworthiness standards as regards the collection and exploitation of operational data~~reliability standards, active monitoring and supervision as regards the collection and exploitation of operational data; AI systems and applications intended to extract and synthesise data, and extrapolate results therefrom to inform decisions for matters relating to defence and national security, must be specific in scope and comply with the provisions set out in the current regulatory framework in terms of gathering and processing data; stresses that AI applications designed to process data for intelligence purposes in defence related activities should comply with data processing standards to avoid risks of unintended surveillance or infringement of individual rights; believes that for high-risk applications of AI-enabled technologies like facial recognition which lack a definitive regulatory framework at the EU level, the Union must ensure that their development and deployment is rightful, proportional and respects the rights of individuals; stresses the importance of monitoring competent national law enforcement authorities which develop and deploy AI-enabled systems and technologies to maintain public order so as to mitigate the disproportional risks of predictive policing;

Amendment 58 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 6

Amendment 59 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 6

Amendment 6 #

Draft opinion
Paragraph 1 a (new)

1a. Welcomes the endorsement, by the 2019 Meeting of High Contracting Parties to the United Nations Convention on Certain Conventional Weapons (CCW), of 11 Guiding Principles for the development and use of autonomous weapons systems;

Amendment 60 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 6

6. Recalls that most of current military powers worldwide have already engaged in significant R&D efforts related to the military dimension of AI; considers that the EU must see to it that it does not lag behind in this regard; Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible trustworthiness standards as regards the collection and exploitation of operational data;

Amendment 61 #

6. Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards to ensure their resilience against vulnerabilities that can exploited by external attacks, cyber- attacks and digital influence targeting the data, the model or the underlying infrastructure, both software and hardware, as well as their compliance with the highest possible trustworthiness standards as regards the collection and exploitation of operational data throughout a system’s entire lifecycle;

Amendment 62 #

Thierry MARIANI, Jérôme RIVIÈRE, Nicolas BAY

Draft opinion
Paragraph 6

6. Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards, according to the principle of "Security by Design", to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible trustworthiness standards as regards the collection, storage and exploitation of operational data;

Amendment 63 #

Draft opinion
Paragraph 6

6. Stresses that for any defence application of AI enabled systems, the ~~Union~~Member States should set technical and organisational standards to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible trustworthiness standards as regards the collection and exploitation of operational data;

Amendment 64 #

Draft opinion
Paragraph 6 a (new)

6a. Stresses that defence applications of AI during operations means that autonomous software in compliance with future EU regulations, can be used to assist with maintenance, logistics management, and targeting of offensive and defensive systems; this could serve to ensure that a CSDP force is successfully integrated with autonomous, unmanned ground, air and sea vehicles to provide a standardised level of operational competence and consistency of execution;

Amendment 65 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 6 a (new)

6a. Notices the great risk of AI in the area of disinformation; underlines that, if not regulated, AI technologies might have ethically adverse effects by exploiting bias in data and algorithms that may lead to disinformation, creating information bubbles and exploiting biases incorporated into AI algorithms;

Amendment 66 #

Draft opinion
Paragraph 6 a (new)

6a. Calls on the Commission and the Member States to work together in establishing a framework to prohibit the activation of AI-enabled systems in the field of security and defence using advanced data to carry out lethal or destructive actions without a human-in- the-loop-principle;

Amendment 67 #

Miriam LEXMANN, Tomáš ZDECHOVSKÝ, Michaela ŠOJDROVÁ, Isabel WISELER-LIMA, François-Xavier BELLAMY, Andrius KUBILIUS, Lukas MANDL

Draft opinion
Paragraph 7

7. Highlights the need to adopt clear reliability, safety and security provisions and requirements for AI-systems in defence, ~~and~~to introduce transparency criteria in the various phases (design, production, operation) and to carry ourt regular tests and verifications across the entire life cycle; underlines the necessity of ensuring compliance with applicable standards and obtained certifications where AI modifies e.g. through machine learning the functionality and behaviour of systems in which it is integrated, in order to ensure full traceability of decisions made with involvement of AI as well as meaningful human control when such systems could kill humans;

Amendment 68 #

Sven MIKSER, Raphaël GLUCKSMANN, Javi LÓPEZ, Nacho SÁNCHEZ AMOR, Eva KAILI, Claudiu MANDA, Giuliano PISAPIA, Attila ARA-KOVÁCS, Andreas SCHIEDER, Juozas OLEKAS, Isabel SANTOS, Tonino PICULA

Draft opinion
Paragraph 7

7. Highlights the need to adopt clear safety and security provisions and requirements, with proper certifications, for AI-systems in defence, and carry ourt constant monitoring, regular tests and verifications across the entire life cycle; underlines the necessity of ensuring compliance with applicable standards and obtained certifications where AI modifies e.g. through machine learning the functionality and behaviour of systems in which it is integrated, in order to ensure full traceability of decisions made with involvement of AI;

Amendment 69 #

Draft opinion
Paragraph 7

7. Highlights the need to adopt clear safety and security provisions and requirements for AI-systems in security and defence, and carry our regular tests and verifications across the entire life cycle; underlines the necessity of ensuring compliance with applicable standards and obtained certifications where AI modifies e.g. through machine learning the functionality and behaviour of systems in which it is integrated, in order to ensure full traceability, explicability and accountability of decisions made with involvement of AI;

Amendment 7 #

Draft opinion
Paragraph 2

2. Stresses that a Union framework regulating the use of artificial intelligence (AI)-enabled systems in defence must respect all applicable legal regimes, in particular the international humanitarian law and the international human rights law, and be in compliance with Union law, principles and values; stresses that the EU should play a global role in leading the way towards a credible and binding AI regulatory agenda rooted in democratic values; calls on the Union to assess the inherent AI-related risks with regard to the application of Union law, and foresee necessary adjustment and enforcement where needed; underlines that emerging technologies not covered by international law should be judged by the principle of humanity and the dictates of public conscience; underlines that the ethics of AI-enabled systems in defence must be assessed from the point of view of Human rights, and notably human safety, health and security, freedom, privacy, integrity and dignity;

Amendment 70 #

Draft opinion
Paragraph 7 a (new)

7a. Calls on the Commission to embed cybersecurity capacity-building in its industrial policy in order to ensure the development and deployment of safe, resilient and robust AI-enabled and robotic systems; calls on the Commission to explore the use of blockchain-based cybersecurity protocols and applications to improve the resilience, trust and robustness of AI infrastructures through disintermediated models of data encryption; encourages European stakeholders to research and engineer advanced features that would facilitate the detection of corrupt and malicious AI- enabled &robotics systems which could undermine the security of the Union and of citizens;

Amendment 71 #

Draft opinion
Paragraph 7 a (new)

7a. Stresses that a digitally shared view of an operational environment, can be developed in seconds to aid overloaded human analysts, and that data sources can be expanded beyond conventional defence-related sources to include open- source and commercially available imagery, metadata, and social media, providing decision makers with a full ‘operating picture’ of a combat or crisis environment;

Amendment 72 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 7 a (new)

7a. Highlights the fact that AI framework in defence and security should develop benchmarks for ethically responsible and accepted uses of AI technologies; underlines that these criteria must be constantly adjusted to the progress in AI technologies;

Amendment 73 #

Draft opinion
Paragraph 7 a (new)

7a. Underlines that AI learning techniques can pose various risks from an ethical point of view, for example, if the underlying data is biased due to ethnically biased population data or the deliberate alteration of learning data by a third party;

Amendment 74 #

Draft opinion
Paragraph 7 b (new)

7b. Emphasises the need for an embedded option to reduce an AI-enabled system’s computations to a form comprehensible by humans throughout a given system’s lifecycle; considers that AI-enabled systems, products and technology purposed for military use should be equipped with a ‘black box’ to record every data transaction carried out by the machine, including the logic that contributed to its decisions, as well as with a 'switch-off' button which would instantly enable humans to deactivate the AI-enabled system;

Amendment 75 #

Draft opinion
Paragraph 7 b (new)

7b. Highlights the fact that European legislation must be flexible and apply to any future technological advancements, and hence must prevent legal loopholes or grey zones as it was sometimes the case in the past;

Amendment 76 #

Draft opinion
Paragraph 8

8. Stressed that all AI-systems in defence must have a concrete and well- defined domain of use and must be endowed with the ability to detect and disengage or diseactivate deployed systems should they move from their domain of use or engage in any escalatory or unintended action; recognises that during operations, machine-learning systems can use sensor data and advanced models to accurately predict and then prevent equipment failure, aiding the ability to prevent and withstand operational failures or logistical shortfalls;

Amendment 77 #

Draft opinion
Paragraph 8

8. Stressed that all AI-systems in defence must have a concrete and well- defined domain of use and must be endowed with the ability to detect and disengage or diseactivate deployed systems should they move from their domain of use or engage in any escalatory or unintended action; Is of the opinion that advanced weaponised AI systems and technology should be prohibited from taking any lethal actions using advanced data stemming from prediction of potential hostile behaviour, without human oversight and confirmation;

Amendment 78 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 8

8. Stressed that all AI-systems in defence must have a concrete and well- defined domain of use and must be endowed with the ability to detect and disengage or diseactivate deployed systems should they move from their domain of use or engage in any escalatory or unintended action; supports the statements of the world’s most prominent AI researchers in their Open Letter from 2015, which calls for a 'ban on offensive autonomous weapons beyond meaningful human control';

Amendment 79 #

Draft opinion
Paragraph 8

8. Stressed that all AI-systems in defence must have a concrete and well- defined ~~domain of use~~mission framework and must be endowed with the ability to detect and disengage or diseactivate deployed systems should they move ~~from their domain of use~~beyond the mission framework defined and assigned by the human command or engage in any escalatory or unintended action;

Amendment 8 #

Draft opinion
Paragraph 2

2. Stresses that a Union framework regulating the use of artificial intelligence (AI)-enabled systems in defence, both in combat and noncombat situations, must respect all applicable legal regimes, in particular the international humanitarian law and the international human rights law, and be in compliance with Union law, principles and values; calls on the Union to assess the inherent AI-related risks with regard to the application of Union law, and ~~foresee~~suggest necessary adjustment and enforcement where ~~needed~~and when needed; keeping in mind the discrepancies in terms of technical and security infrastructures throughout the European Union;

Amendment 80 #

Draft opinion
Paragraph 8

Amendment 81 #

Draft opinion
Paragraph 8

8. Stresseds that all AI-systems in defence must have a concrete and well- defined domain of use ~~and must be endowed with~~whereby humans retain the a~~bilit~~gency to detect ~~and~~, disengage or diseactivate deployed systems should they move from their domain of use or engage in any escalatory or unintended action;

Amendment 82 #

Draft opinion
Paragraph 8 a (new)

8a. Highlights the importance of ensuring that AI-enabled systems, weapons, products and technology that are produced in the Union have advanced software security provisions in accordance with the "safety by design approach" which would render it difficult to hack or interfere with by foreign stakeholders or malicious groups, and allow for specific human oversight before operating such systems compromised by unknown sources;

Amendment 83 #

Draft opinion
Paragraph 8 a (new)

8a. Stresses, especially in security and defence domains, the importance of human oversight over AI, which can be achieved through a human-in-command approach, meaning the capability by human operators to oversee the overall activity of the AI system and the ability to decide when and how to use the system in any particular situation, including the ability to override a decision made by the system at any time;

Amendment 84 #

Draft opinion
Paragraph 8 b (new)

8b. Recalls its previous positions on lethal autonomous weapon systems and calls once again for the urgent development and adoption of a common position on lethal autonomous weapon systems, for an international ban on the development, production and use of lethal autonomous weapon systems enabling strikes to be carried out without meaningful human control, and for a start to effective negotiations for their prohibition;

Amendment 85 #

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators ~~and~~, as there must be meaningful human control over any weapons system and human intent in the decision to use force; underlines that the human-in- the-loop principle must also be applied to the command and control of AI- enabled systems; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and ~~exercise the necessary level of judgment~~accountability for the use of lethal force and exercise the necessary level of judgment, which cannot be endowed to machines as it must be based on distinction, proportionality and precaution, for taking lethal or large-scale destructive action bey means of such systems; recalls in this respect its position on a ban on the development, production and use of fully autonomous weapons systems enabling strikes to be carried out without meaningful human intervention;

Amendment 86 #

Anna BONFRISCO, Marco ZANNI, Susanna CECCARDI, Lars Patrick BERG

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop principle must also be applied to the command and control of AI-enabled systems; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems; emphasises the need for humans to always retain the final control in the execution of any decision of AI empowered weapons that might have lethal consequences; stresses the need to establish clear and traceable authorisation and accountability frameworks for the deployment of smart weapons and other AI-enabled systems, using unique user characteristics like biometric specifications to enable deployment exclusively by authorised personnel;

Amendment 87 #

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop and human-on-the-loop principles must also be applied to the command and control of AI-enabled systems depending on the nature of the system and of the mission at hand and without compromising the efficiency of the application; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems;

Amendment 88 #

Kinga GÁL

Draft opinion
Paragraph 9

9. UEmphasizes that military robots are becoming more complex; underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop principle must also be applied to the command and control of AI-enabled systems; stresses that AI- enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems, such decisions must be made by human operators;

Amendment 89 #

Draft opinion
Paragraph 9

Amendment 9 #

Robert BIEDROŃ, Javi LÓPEZ, Sven MIKSER, Andreas SCHIEDER

Draft opinion
Paragraph 2

Amendment 90 #

Nathalie LOISEAU, Klemen GROŠELJ, Javier NART, Bernard GUETTA, Petras AUŠTREVIČIUS, Christophe GRUDLER, Bart GROOTHUIS

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop, human-on-the-loop and human- in-command principles must also be applied to the command and control of AI- enabled systems especially when it comes to conflicts of objectives; stresses that AI- enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action bey means of such systems;

Amendment 91 #

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human~~- in-the-loop principle must also be applied to the command and control of AI-enabled systems~~ control should remain effective for the command and control, following a human- on-the-loop principle at the military leadership level; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems;

Amendment 92 #

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop and human-in-command principles must also be applied to the command and control of AI-enabled systems; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems;

Amendment 93 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS, Dragoş TUDORACHE

Draft opinion
Paragraph 9

9. Underlines that the entire responsibility and accountability for the decision to design, develop, deploy and use AI-systems must rest on human operators and the human-in- the-loop principle must also be applied to the command and control of AI-enabled systems; stresses that AI- enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgment for taking lethal or large-scale destructive action be means of such systems;

Amendment 94 #

Draft opinion
Paragraph 9 a (new)

9a. Stresses that the EU must take the lead in promoting the establishment of international norms regarding the ethical and legal parameters of the development and use of fully autonomous, semi- autonomous and remotely operated lethal weapons systems; Member States should develop national strategies for the definition, status and use of lethal autonomous weapons (LAWs) towards a comprehensive strategy on the EU level;

Amendment 95 #

Draft opinion
Paragraph 9 a (new)

9a. Stresses that in tactical scenarios, AI techniques such as reinforcement learning, which allows machines to share their experiences and optimal solutions among themselves, have proven to be a critical asset in military campaigns, leading to the evolution of a highly optimised, robust mission intelligence that is effective at fulfilling objectives set by military command;

Amendment 96 #

Draft opinion
Paragraph 9 a (new)

9a. Stresses the importance of putting in place mechanism to ensure responsibility and accountability for AI systems and their outcomes, including the setup of governance frameworks ensuring accountability for the ethical dimensions of decisions associated with the development, deployment and use of AI;

Amendment 97 #

Draft opinion
Paragraph 9 a (new)

9a. Welcomes the agreement of Council and Parliament to exclude lethal autonomous weapons ‘without the possibility for meaningful human control over the selection and engagement decisions when carrying out strikes’ from actions funded under the European Defence Fund;

Amendment 98 #

Draft opinion
Paragraph 9 b (new)

9b. Calls on the Commission to work together with Member States’ national competent authorities and other stakeholders participating in the development and deployment of AI- enabled systems, products and technologies to establish a safe, secure and resilient framework whereby the source code of AI-enabled systems is shared, monitored and verified to mitigate potential deviations from the governing principles and ethical framework underpinning AI technology in the field of security and defence; suggests to the Commission that the EU must retain ownership of the intellectual property of EU-funded research on AI-enabled systems, products and technologies in security and defence;

Amendment 99 #

Katalin CSEH, Klemen GROŠELJ, Urmas PAET, Hilde VAUTMANS, Petras AUŠTREVIČIUS, Dragoş TUDORACHE, Nathalie LOISEAU

Draft opinion
Paragraph 10

10. Underlines that the Union must promote understanding the military implications of AI, of robotics and of autonomy; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI-enabled military capabilities; underlines the urgent need for establishing increased European strategic and technological independence in the field of AI enabled systems, including the critical infrastructure it relies on;

source: 650.709

2020/05/19 TRAN 200 amendments...

Amendment 1 #

Draft opinion
Recital A

A. whereas Artificial Intelligence (AI) is a strategic technology for the transport sector ~~and is expect~~that needs to benefit all citizens and society, by improving the quality of life, creating new employment opportunities and more sustainable business models and needs to be designed to be widely affordable;

Amendment 1 #

Draft opinion
Citation 1 a (new)

- Whereas there are serious concerns that the current Union legal framework, including the consumer law acquis, data protection legislation, product safety and market surveillance legislation, is not fit for purpose to effectively tackle the risks created by artificial intelligence, robotics and related technologies and does not provide a high level of consumer protection as required in Article 38 of the Charter of Fundamental Rights of the European Union;

Amendment 10 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Marc ANGEL, Adriana MALDONADO LÓPEZ, Andreas SCHIEDER

Draft opinion
Recital D

D. whereas the EU aimed to reduce annual road fatalities in the EU by 50% by 2020 compared to 2010, but, in view of stagnating progress, renewed its efforts in its Road Safety Policy Framework 2021 - 2030 - Next steps towards "Vision Zero"; whereas in this regard, automation and new technologies have an essential role and a great potential to increase road safety by avoiding human error;

Amendment 10 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Jordi CAÑAS, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 2

2. Notes that the framework should apply to algorithmic systems, including the fields of artificial intelligence, machine learning, deep learning, automated and assisted decision making processes and robotics;

Amendment 100 #

Draft opinion
Paragraph 10

10. Calls for the Union to consider the establishment of a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities; further calls for this structure to be appropriately advised by stakeholder organisations (such as consumer protection organisations) in order to ensure wide consumer representation;

Amendment 101 #

Maria da Graça CARVALHO, Edina TÓTH, Andrey KOVATCHEV, Marion WALSMANN

Draft opinion
Paragraph 10

10. Calls ~~for~~on the UnCommission to establish a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities; considers that due to the disproportionate impact of algorithmic systems on women and minorities, the decisional levels of such a structure should be diverse and gender balanced;

Amendment 102 #

Christel SCHALDEMOSE, Leszek MILLER, Evelyne GEBHARDT, Marc ANGEL, Alex AGIUS SALIBA, Sándor RÓNAI, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Andreas SCHIEDER, Sylvie GUILLAUME, Brando BENIFEI

Draft opinion
Paragraph 10

10. Calls for the Union to establish a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities; emphasizes that Member States must develop risk-management strategies for AI in the context of their national market surveillance strategies.

Amendment 103 #

Jordi CAÑAS, Svenja HAHN

Draft opinion
Paragraph 10

10. Calls for the ~~Union to establish a European market surveillance structure for algorithmic systems~~Commission to promote the exchange of information related to algorithmic systems between market surveillance authorities, and to support the development of a common understanding in the Single Market by issuing guidance, opinions and expertise to Member States’ authorities;

Amendment 104 #

Draft opinion
Paragraph 10

10. Calls for the Union to establish a European market surveillance structure ~~for algorithmic system~~composed of national market surveillance authorities issuing guidance, opinions and expertise to Member States’ authorities;

Amendment 105 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 10

10. Calls for the ~~Union to establish a European~~use of existing bodies on either Union or member state level to conduct market surveillance ~~structure~~ for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities;

Amendment 106 #

Draft opinion
Paragraph 10 a (new)

10a. Calls on the Commission and the authorities of the Member States to combat tax evasion and the abuse of a dominant position by non-European companies in Europe active in the market for algorithmic systems;

Amendment 107 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Sándor RÓNAI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 11

11. Notes that as regards the high risk applications, it is essential for the software documentation, the algorithms and data sets used to be fully accessible to market surveillance authorities designated by Member States, while respecting Union law~~; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect~~;

Amendment 108 #

Draft opinion
Paragraph 11

11. Notes that it is essential for the software documentation, the algorithms and data sets used or produced by artificial intelligence, robotics, and related technologies to be fully accessible to market surveillance authorities, while respecting Union law; further notes that such elements should be preserved by those who are involved in the different stages of the development of algorithmic systems and in proportion of their liability, namely through distributed ledger technologies, such as block-chain; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect;

Amendment 109 #

Draft opinion
Paragraph 11

11. Notes that ~~it is essential for~~ the software documentation, ~~the~~ algorithms and data sets ~~used to be fully~~should be explainable and accessible to market surveillance authorities~~, whil~~ to the gre~~specting Union law~~atest possible extent, in accordance with Union law and particularly a future regulatory framework of ethical aspects; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect; considers that an examination of the current market surveillance legislation might be necessary to avoid its obsolescence and ensure that it responds ethically to the emergence of artificial intelligence, robotics and related technologies;

Amendment 11 #

Draft opinion
Recital D a (new)

D a. whereas human error is involved in about 95% of all road traffic accidents in the EU. Driverless cars and lorries can drastically reduce these figures and improve road safety, while new digital technologies can also reduce traffic congestion and emissions of greenhouse gases and air pollutants

Amendment 11 #

Maria da Graça CARVALHO, Edina TÓTH, Róża THUN UND HOHENSTEIN, Marion WALSMANN, Pascal ARIMONT

Draft opinion
Paragraph 2

2. Notes that the regulatory framework should apply to algorithmic systems, including the fields of artificial intelligence, machine learning, deep learning, automated decision making processes and robotics;

Amendment 110 #

Draft opinion
Paragraph 11

11. Notes that it is essential for the software documentation, the algorithms and data sets used to be ~~fully accessi~~explainable to market surveillance authorities, while respecting Union law; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect;

Amendment 111 #

Draft opinion
Paragraph 11

11. Notes that it is essential for the software documentation, the algorithms and data sets used to be ~~fully~~ accessible to market surveillance authorities, while respecting Union law and trade secrets; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect;

Amendment 112 #

Draft opinion
Paragraph 11

11. Notes that it is essential for the risk assessment documentation, the software documentation, the algorithms and data sets used to be fully accessible to market surveillance authorities, while respecting Union law; ~~invi~~notes th~~e Commission to assess if~~at additional prerogatives should be given to market surveillance authorities in this respect;

Amendment 113 #

Draft opinion
Paragraph 12

12. Calls for the designation and sufficient funding by each Member State of a competent national authority for monitoring the application of the provisions related to algorithmic systems;

Amendment 114 #

Draft opinion
Paragraph 12

12. Calls for the designation by each Member State of a competent national authority or of several competent regional authorities for monitoring the application of the provisions;

Amendment 115 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Evelyne GEBHARDT, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 12

12. Calls for the designation by each Member State of a competent national authority for monitoring the application of the provisions related to algorithms;

Amendment 116 #

Draft opinion
Paragraph 12

12. Calls for the designation by each Member State of a competent national ~~authority~~enforcement body(NEB) for monitoring the application of the provisions;

Amendment 117 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 12 a (new)

12a. Stresses the need for the authorities to put an end to the practices of massive and systematic profiling or tracking of consumers organised by some players in the advertising industry on the internet or mobile phones; notes that these tools collect very sensitive data about behaviour, interests or health, even from minors and that they are then used by third parties; regrets that the lack of supervision by the authorities strengthens the market power of less ethical players and dissuades some players from offering more ethical solutions that are more favourable to consumers.

Amendment 118 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 13

13. Calls for the establishment of a European market surveillance board for algorithmic systems, to ensure a level playing field and to avoid fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national authorities;deleted

Amendment 119 #

Draft opinion
Paragraph 13

Amendment 12 #

Draft opinion
Recital D a (new)

D a. Whereas the importance of an ethical and regulatory framework is recognised as crucial, it also needs to include provisions on the quality of data sets used in algorithmic systems, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards;

Amendment 12 #

Draft opinion
Paragraph 2

2. Notes that the framework should apply to algorithmic systems, including the fields of artificial intelligence, machine learning, ~~deep learning~~rule-based systems, automated decision making processes and robotics;

Amendment 120 #

Draft opinion
Paragraph 13

13. Calls ~~for~~on the establishment of a European market surveillance board for algorithmic systems, to ensure a level playing field and to avoid fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national authoritiesCommission to ensure that Union legislation related to algorithmic systems is enforced in all Member States to avoid fragmentation of the internal market;

Amendment 121 #

Maria da Graça CARVALHO, Róża THUN UND HOHENSTEIN, Edina TÓTH, Marion WALSMANN

Draft opinion
Paragraph 13

13. Calls for ~~the establishment of a European market~~improved and effective cooperation of national surveillance bo~~ard for~~dies as regards algorithmic systems~~, to ensure a level playing fi~~ across the Union, as weldl a~~nd to avoid fr~~s harmonised risk management~~ation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used~~ strategies for artificial intelligence, in mor~~e than one Member State, as w~~der to ensure a level playing fielld a~~s at the request of the majority of the national authorities~~nd to avoid fragmentation of the internal market;

Amendment 122 #

Draft opinion
Paragraph 13

Amendment 123 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Jordi CAÑAS, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 13

13. Calls for ~~the establishment of~~a strong coordination of Member State’s authorities, for instance through instruments such as a European market surveillance board for algorithmic systems, to ensure a level playing field and to avoid fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national authorities;

Amendment 124 #

Draft opinion
Paragraph 13

13. Calls for the establishment of a European market surveillance ~~board~~advisory committee for algorithmic systems, ~~to ensure~~composed of the representatives of the national competent authorities, the Commission and relevant EU agencies such as ENISA, to contribute to a level playing field and to avoiding fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national by issuing guidelines and recommendations to Member States’ authorities;

Amendment 125 #

Evelyne GEBHARDT, Marc ANGEL, Monika BEŇOVÁ, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME, Sándor RÓNAI, Brando BENIFEI

Draft opinion
Paragraph 13

13. Calls for the establishment of a European market surveillance board for algorithmic systems in the framework of a European Agency for Artificial Intelligence, to ensure a level playing field and to avoid fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national authorities;

Amendment 126 #

Draft opinion
Paragraph 13

Amendment 127 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Evelyne GEBHARDT, Marc ANGEL, Alex AGIUS SALIBA, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Sándor RÓNAI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 13 a (new)

13a. Acknowledges valuable outputs of the High-Level Expert Group on Artificial Intelligence, particularly ‘The Ethics Guidelines for Trustworthy Artificial Intelligence’; suggests that this group comprising representatives from academia, civil society and industry, as well as the European AI Alliance, might provide expertise to the European market surveillance advisory committee for algorithmic systems;

Amendment 128 #

Evelyne GEBHARDT, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME, Sándor RÓNAI

Draft opinion
Paragraph 13 a (new)

13a. Believes that a European certification of ethical compliance should be designed in such a way as to inform consumers about the risk level of a product or a service with an algorithmic component as well as its trustworthiness in the light of the ethical principles and all other requirements based on relevant Union legislation;

Amendment 129 #

Draft opinion
Paragraph 13 a (new)

13a. Stresses the need for national market surveillance authorities to be reinforced in terms of capacity, skills, and competences in AI, IoT and other related technologies, as well as knowledge about its specific risks;

Amendment 13 #

Tom BERENDSEN, Marian-Jean MARINESCU, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE

Draft opinion
Recital D a (new)

D a. whereas the production of ethically responsible, human-centred and technologically robust AI, robotics and related technologies in transport present European businesses, including SMEs, a business opportunity to become global leaders in this area; whereas such opportunity is particularly present, considering the current global leaders in this area produce technologies that are deemed insufficiently trustworthy and not adequately respecting ethical principles.

Amendment 13 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 2 a (new)

2a. Believes that the Commission should complete a full review of the existing legislation in order to identify legislative gaps; underlines in that regard the extensive legislation already in force that guarantees for instance that products and services placed on the Union market are safe and do not harm people, respect their privacy and follow stringent environmental rules; calls on the Commission to refrain from adopting a legislative act that would double, overlap or contradict those sector-specific legislations;

Amendment 130 #

Virginie JORON, Jean-Lin LACAPELLE, Marco CAMPOMENOSI

Draft opinion
Paragraph 13 a (new)

13a. Demands an update of European competition and public procurement law to promote the emergence of world-class players based in Europe;

Amendment 131 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 13 b (new)

13b. Calls for measures to encourage global companies to relocate part of their employment, research and digital systems and products to Europe, if they want to benefit from full access to the European market;

Amendment 132 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 13 c (new)

13c. Calls for respect for European and national preferences, for the development of European territories and employment in the sector of artificial intelligence and robotics; stresses the importance of preventing the takeover of strategic companies by non-European players;

Amendment 133 #

Draft opinion
Subheading 5a (new)

Equality and non-discrimination

Amendment 134 #

Draft opinion
Paragraph 13 d (new) (after Subheading 5 a new)

13d. Highlights the need for equal respect for the moral worth and that dignity of all human beings must be ensured. This goes beyond non- discrimination, which tolerates the drawing of distinctions between dissimilar situations based on objective justifications. In an AI context, equality entails that the system’s operations cannot generate unfairly biased outputs (e.g. the data used to train AI systems should be as inclusive as possible, representing different population groups).; calls for adequate protection for potentially vulnerable persons and groups, such as workers, women, persons with disabilities, ethnic minorities, elderly people, children, consumers or others at risk of exclusion;

Amendment 135 #

Draft opinion
Paragraph 13 e (new) (after Subheading 5 a new)

13e. Notes that, particularly in business-to-consumer domains, systems should be user-centric and designed in a way that allows all people to use AI products or services, regardless of their age, gender, abilities or characteristics. Accessibility to this technology for persons with disabilities, which are present in all societal groups, is of particular importance. AI systems should not have a one-size-fits-all approach and should consider Universal Design principles addressing the widest possible range of users, following relevant accessibility standards. This will enable equitable access and active participation of all people in existing and emerging computer-mediated human activities and with regard to assistive technologies;

Amendment 136 #

Draft opinion
Paragraph 14 a (new)

14a. The main principles and aims of the proposal are: 1. A genuine Digital Single Market requires full harmonisation by a Regulation. 2. For algorithmic systems that are considered of a higher risk, horizontally applicable safeguards are essential, irrespective of sector-specific rules. 3. The horizontal Regulation should build on and ensure the implementation of the European framework of rights of users and consumers, in particular the protection of privacy, non-discrimination, dignity, fairness, freedom of expression and the right to an effective remedy.

Amendment 137 #

Maria da Graça CARVALHO, Róża THUN UND HOHENSTEIN, Edina TÓTH, Marion WALSMANN

Draft opinion
Paragraph 14 b (new)

14b. The determination of the risk of algorithmic systems in a future horizontal regulation on artificial intelligence should be based on a combination of the severity of the potential damage and the probability of its occurrence. An increased risk potential of an algorithmic system must be accompanied by a higher degree of regulatory intervention. For the algorithmic systems in the lowest risk category, no further legal obligations should apply. A pre-defined set of criteria should be used to define the risk potential of an application or a scope of application. These criteria should include: 1. The quality and integrity of the collected and processed data 2. The intensity of the possible violation of rights, by taking into account: (a) the depth of the potential damage caused; (b) the number of persons affected; (c) the sum of potential damages, since on the one hand, there may be particularly intensive individual cases of rights violations and, on the other hand, there may be particularly frequent minor rights violations whose sum nevertheless has social relevance. 3. The impact on fundamental rights and freedoms; 4. The impact on society and the environment, including: (a) material/monetary damage and welfare effects for individuals or groups, for example in the allocation of resources or access to markets; (b) social implications, such as social, psychological, cultural and economic dimensions; (c) the systemic relevance for the functioning of society and democracy, such as elections, the formation of public opinion or the creation of socially useful infrastructure; (d) the impact on environmental sustainability, such as its energy consumption, the use of raw materials or effects on the climate. 5. The likelihood of the occurrence of damage including: (a) the role of the algorithm in taking the decision; (b) the dependence of the persons impacted, for example private sector services in competitive environments versus public services provided by the state or services by a market dominant company; (c) the reversibility of the consequences of a decision.

Amendment 14 #

Jörgen WARBORN

Draft opinion
Recital D a (new)

Da. Drawing attention to the great opportunities that exist to make greater use of AI technology in the transport industry, and whereas, in the light of this, future regulation must be so designed as to facilitate technological development rather than hampering it;

Amendment 14 #

Draft opinion
Paragraph 2 a (new)

2a. Stresses that an EU regulatory framework of AI shall have a human- centric approach and lead to development of systems which incorporate European ethical values by-design; considers that an EU regulatory framework that focuses on European values would be an added value providing Europe with a unique competitive advantage and make a significant contribution to the well-being and prosperity of European citizens and businesses, and boost our internal market;

Amendment 15 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Recital D b (new)

D b. whereas such new business opportunities may contribute to the recovery of the European industry after the current health and economic crisis; whereas such opportunities will create new jobs as the uptake of AI and related technologies has the potential to increase businesses' productivity levels and contribute to efficiency gains; whereas innovation programs in this area can enable regional clusters to thrive.

Amendment 15 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 2 b (new)

2b. Points out that the legislative framework introduced by Decision No 768/2008/EC1a provides for a harmonised list of obligations for producers, importers and distributors, encourages the use of standards and foresees several levels of control depending on the dangerousness of the product; considers that this framework should also apply to AI imbedded products; __________________ 1aDecision No 768/2008/EC of the European Parliament and of the Council of 9 July 2008 on a common framework for the marketing of products, and repealing Council Decision 93/465/EEC (OJ L 218, 13.8.2008, p. 82).

Amendment 16 #

Jörgen WARBORN

Draft opinion
Recital D b (new)

Db. calling on the Commission to take measures to improve Europe's competitiveness in the field of AI, with the aim of creating a climate for innovation in which European undertakings and organisations can become world leaders in the development of AI technology;

Amendment 16 #

Maria da Graça CARVALHO, Edina TÓTH

Draft opinion
Paragraph 2 b (new)

2b. Underlines that an ethical framework of AI is an added value to promote innovation on the market;

Amendment 17 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Recital D c (new)

D c. whereas such European approach to the development of AI, robotics and related technologies in transport has the potential to increase the global competitiveness and strategic autonomy of the European economy.

Amendment 17 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 2 c (new)

2c. Considers that for the future legislation to apply, legal obligations need to be very precise and avoid to refer to general principles in order to ensure that they are implementable by economic operators;

Amendment 18 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ

Draft opinion
Paragraph 1

1. ~~Highlight~~Understands the potential of AI for all autonomous means of road, rail, waterborne and air transport, but keeps in mind that an ethical approach to AI can also be seen as an early warning system against risks;

Amendment 18 #

Draft opinion
Paragraph 2 d (new)

2d. Considers that mandatory rules as regards the regulatory framework of ethical aspects should be limited to practices that would undoubtedly undermine fundamental rights and freedoms;

Amendment 19 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ

Draft opinion
Paragraph 1

1. Highlights the potential of AI for safety and efficiency on all autonomous means of road, rail, waterborne and air transport;

Amendment 19 #

Draft opinion
Paragraph 3

3. Stresses that any future regulation of ethical aspects should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk categoryies there should be no additional legal obligations; ~~algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this~~the risk-based approach should follow clear and transparent rules;

Amendment 2 #

Draft opinion
Recital A

A. whereas Artificial Intelligence (AI) is a strategic technology for the transport sector and is expected to benefit citizens and society, by improving the quality of life, raising the safety level of all modes of transport, creating new employment opportunities and more sustainable business models;

Amendment 2 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Citation 1 b (new)

- Whereas ethical guidance, such as the principles adopted by the High-Level Expert Group on Artificial Intelligence, provides a good starting point but is not enough to ensure that businesses act fairly and guarantee effective consumer protection;

Amendment 20 #

Draft opinion
Paragraph 1

1. Highlights the potential of AI, robotics and related technologies for all autonomous means of road, rail, waterborne and air transport;

Amendment 20 #

Maria da Graça CARVALHO, Edina TÓTH, Andrey KOVATCHEV

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, with clear criteria and indicators, followed by an impartial and regulated assessment based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no additional legal obligations; ~~algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category;~~ this risk-based approach should follow by clear and transparent rules;

Amendment 21 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Marc ANGEL, Adriana MALDONADO LÓPEZ, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 1

1. Highlights the potential of AI for all autonomous means of road, rail, waterborne and air transport of passengers and goods;

Amendment 21 #

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm or breaches of rights for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations and certification requirements should gradually increase with the identified risk level; in the lowest risk category there should be no additional legal obligations~~; algorithmic systems that may harm an individual~~ and certification or labelling systems should be voluntary; algorithmic systems that may harm an individual or cause potential breaches of an individual’s rights, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules, providing enough legal certainty whilst being future-proof;

Amendment 22 #

Draft opinion
Paragraph 1

1. Highlights the potential of AI for all autonomous means of road, rail, waterborne and air transport; considers that IA should also incorporate blockchain and distributed ledger technologies (DLT) due to their importance in the transaction, communication and information-sharing;

Amendment 22 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no additional legal obligations, whereas applications in the highest risk category should be deemed illegal; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules; the risk assessment of a specific system must be subject to regular re-evaluation;

Amendment 23 #

Draft opinion
Paragraph 1 a (new)

1 a. Moreover, highlights the potential of AI, robotics and related technologies to boost the modal shift and intermodality, as these technologies can contribute to finding the optimal combination of transport modes for the transport of goods and passengers

Amendment 23 #

Draft opinion
Paragraph 3

3. Stresses that any possible future regulation should enable development and deployment of secure and trustworthy algorithmic systems and that it should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified basic, substantial or high risk level; in the lowest risk category there should be no additional legal obligations; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules;

Amendment 24 #

1 a. Highlights that AI should be designed and developed in ways that decrease inequality and further social empowerment, with respect for human autonomy, and increase benefits that are shared by all, equitably. It is especially important that AI be explicable, as explicability is a critical tool to build public trust in, and understanding of, the technology;

Amendment 24 #

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should ~~gradual~~only increase ~~with the identified~~in accordance with the proven risk level; in the lowest risk category there should be no additional legal obligations; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules;

Amendment 25 #

Henna VIRKKUNEN

Draft opinion
Paragraph 1 a (new)

1 a. Highlights that the global competition between companies and economic regions requires, that the EU promotes investments and strengthens the international competitiveness of companies operating in the transport sector by establishing itself as an environment favourable for the development and application of AI solutions.

Amendment 25 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for ~~society at large~~the whole society, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no additional legal obligations; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules;

Amendment 26 #

Henna VIRKKUNEN

Draft opinion
Paragraph 1 b (new)

1 b. Stresses that while there is a clear need to update of the regulatory framework concerning these emerging technologies and their use in the transport sector, it must be ensured that such updates are always based on a real need, comply with the principle of better regulation, and will not impose unnecessary regulatory burden on businesses and consumers.

Amendment 26 #

Evelyne GEBHARDT, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME

Draft opinion
Paragraph 3

3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be ~~no additional legal~~a labelling obligations; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk- based approach should follow clear and transparent rules;

Amendment 27 #

Draft opinion
Paragraph 1 b (new)

1 b. Furthermore, stresses that AI, robotics and related technologies have the potential to make transport, logistics and traffic flows more efficient and to make all transport modes safer, smarter, and more environmentally friendly.

Amendment 27 #

Draft opinion
Paragraph 3 a (new)

3a. Believes that prior to tabling possible new legislative proposals, the Commission should scrutinize application of the legislation already in place and its enforcement, as well as self-regulatory measures; stresses that any legislative proposals related to algorithmic systems should not hamper the emergence of high-tech ‘unicorns’, start-ups and SMEs in Europe or prevent European companies from adopting and implementing AI or other solutions and that such proposals should take into consideration the context of the post COVID-19 crisis period; reiterates that the main objective should be to unleash the potential of algorithmic systems throughout the Union and to increase competitiveness of European companies on a global scale;

Amendment 28 #

Draft opinion
Paragraph 2 – introductory part

2. Stresses that the EU transport sector needs a clear framework for achieving trustworthy AI, including safety, security and liability aspects, which will be key to boosting investments in research and innovation, development of skills and the uptake of AI by public services, SMEs, start-ups and businesses and at the same time ensuring data protection as well as interoperability; and in this regard:

Amendment 28 #

Tom BERENDSEN, Marian-Jean MARINESCU, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 3 a (new)

3a. Stresses that a framework of ethical aspects should include provisions serving as a reference for engineers, developers, companies and other operators, with the ultimate aim of promoting responsible and ethical deployment, selling and use of AI, robotics and related technologies, as well as for the development of technical standards and certification procedures in the Union;

Amendment 29 #

Draft opinion
Paragraph 2 – introductory part

2. Stresses that the EU transport sector needs a clear ethical framework for achieving trustworthy AI, including safety, human autonomy and oversight, and liability aspects, which will be key to boosting investments in research and innovation, development of skills and the uptake of AI by SMEs, start-ups and businesses; and in this regard:

Amendment 29 #

Draft opinion
Paragraph 3 a (new)

3a. Calls for the framework to prohibit the deployment or use by Member States of remote recognition technologies, such as biometric recognition that automatically identifies individuals, even for the purpose of responding to a national emergency;

Amendment 3 #

Draft opinion
Recital A

A. whereas Artificial Intelligence (AI) is a strategic technology for the transport sector and is expected to benefit citizens and society, by improving the quality of life, creating new employment opportunities and more sustainable business models; whereas AI has the potential to transform society significantly, particularly in terms of transport and infrastructures; whereas AI should also be a tool to achieve the UN Sustainable Development Objectives;

Amendment 3 #

Tom BERENDSEN, Marian-Jean MARINESCU, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA

Draft opinion
Citation 1 c (new)

- Whereas artificial intelligence, robotics and related technologies may put consumers at risk of being manipulated and subject to discriminatory treatment and arbitrary, intransparent decisions, thereby contributing to increasing asymmetries of power between businesses and consumers, placing consumers at an even more vulnerable position;

Amendment 30 #

Draft opinion
Paragraph 2 – point a

a) calls on the Commission to provide for a liability mechanism, based on a risk- assessment approach, as the impact on the development and uptake of AI of choosing who should be strictly liable for AI- enabled operations of a vehicle needs to be addressed carefully; stresses in this regard that liability should always lie with natural or legal persons;deleted

Amendment 30 #

Evelyne GEBHARDT, Maria-Manuel LEITÃO-MARQUES, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME, Sándor RÓNAI

Draft opinion
Paragraph 3 a (new)

3a. Believes that the ethical principles should be the basis for a harmonised European system of risk classification and related legal obligations;

Amendment 31 #

Tom BERENDSEN, Marian-Jean MARINESCU, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 2 – point a a (new)

a a) calls on the Commission to provide for a clear framework of ethical principles for the development, deployment and use of AI, robotics and related technologies in the transport sector; any AI, robotics and related technologies in the transport sector shall be developed, deployed and used in accordance with those ethical principles.

Amendment 31 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Alex AGIUS SALIBA, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 4

4. Underlines the importance of an ethical and regulatory framework, including in particular provisions on the quality of data sets used in algorithmic systems in relation to the purpose of its use, especially regarding the representativeness of training data used, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards; stresses that those data sets should be auditable and made available to the competent authorities whenever called upon to ensure their fitness with the previously exposed principles;

Amendment 32 #

Draft opinion
Paragraph 2 – point b

b) recommends the establishment of guidelines for a harmonised risk classification of AI-enabled technologies in transport, covering vehicle functions allocated to humans and to AI, and clarifying responsibilities and requirements as regards safety;deleted

Amendment 32 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 4

4. UObserves that availability and quality of data is key to address risks which derive from application of the algorithmic systems; underlines the importance of an ethical and regulatory framework including in particular provisions on the quality of data sets used in algorithmic systems, especially regarding the representativeness of training data used, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards;

Amendment 33 #

Draft opinion
Paragraph 2 – point b

b) recommends the establishment of guidelines for a harmonised risk classification of AI-enabled technologies in all modes of transport, covering vehicle functions allocated to humans and to AI, and clarifying responsibilities and requirements as regards safety;

Amendment 33 #

Draft opinion
Paragraph 4

4. Underlines the importance of ~~an ethical and regulatory framework including in particular~~unlocking large amounts of high quality data to train AI systems; highlights that an ethical regulatory framework should include provisions on the quality of data sets used in algorithmic systems, ~~especially~~for instance regarding the representativeness of training data used, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards;

Amendment 34 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Jörgen WARBORN

Draft opinion
Paragraph 2 – point b a (new)

b a) calls on the Commission to establish a European market surveillance structure for algorithmic systems, including their associated data protection provisions, issuing guidance, opinions and expertise to Member States’ authorities, including on interoperability;

Amendment 34 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Jordi CAÑAS

Draft opinion
Paragraph 4

4. Underlines the importance of a~~n ethical and regulatory framework~~ framework of ethical aspects including in particular provisions on the quality of data sets used in algorithmic systems, especially regarding the representativeness of training data used, on the de-biasing of data sets, ~~as well as on the algorithms themselves,~~ and on data and aggregation standards;

Amendment 35 #

Draft opinion
Paragraph 2 – point c

c) calls on the Commission to set up an AI insurance scheme for intelligent transport systems, in line with the High Level Expert Group risk classification, in order to respond better to the emerging needs of the transport sector;deleted

Amendment 35 #

Maria da Graça CARVALHO, Edina TÓTH, Andrey KOVATCHEV, Róża THUN UND HOHENSTEIN, Pascal ARIMONT

Draft opinion
Paragraph 4

4. Underlines the importance of an ethical and regulatory framework including in particular provisions on the quality of data sets used in algorithmic systems, especially regarding the ~~representativeness of training data used, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards~~high-quality and, where possible, unbiased data sets in order to improve the output of algorithmic systems and boost consumer trust and acceptance;

Amendment 36 #

Jörgen WARBORN

Draft opinion
Paragraph 2 – point c a (new)

(ca) Calls on the Commission to devote particular attention to the situation of SMEs and to design future legislation in such a way as to improve the opportunities for these undertakings to develop and use AI technology.

Amendment 36 #

Draft opinion
Paragraph 4 a (new)

4a. Highlights that user safety, data security, protection of personal data and ethical concerns altogether will determine public acceptance and consequent market penetration of automated systems; highlights that public authorities and private stakeholders will need to provide credible answers to all these concerns as well as prove the environmental, economic, social and safety benefits of AI in order to gain public trust;

Amendment 37 #

Draft opinion
Paragraph 2 – point c a (new)

(ca) considers it necessary to provide detailed information to end- users regarding the operation of transport systems and AI-based vehicles;

Amendment 37 #

Draft opinion
Paragraph 4 a (new)

4a. Points out that a framework of ethical aspects of artificial intelligence, robotics and other technologies should particularly provide for legally binding measures and standards to prevent practices from private and public actors involved that could undermine fundamental rights and freedoms, in order to provide the highest level of consumer protection and to ensure the development of trustworthy, ethical and technically robust applications in the Single Market;

Amendment 38 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 3

3. CHighlights that the European approach to AI technology should secure people’s trust, serve the public interest, and strengthen shared social responsibility and considers the development of trustworthy, ethically responsible and technically robust AI an important enabler for sustainable and smart mobility; in this regard, calls on the Commission to promote uptake of AI in the transport sector and to propose, if necessary in order to uphold EU fundamental rights, corresponding changes to EU legislation without delay and in close cooperation with the transport sector;

Amendment 38 #

Draft opinion
Paragraph 4 a (new)

4a. Stresses that the personal data of European citizens should preferably be processed in Europe.

Amendment 39 #

Draft opinion
Paragraph 3

3. Considers the development of trustworthy, ethically responsible and technically robust AI an important enabler for sustainable, safe, accessible and smart mobility; in this regard, calls on the Commission to ~~promot~~continue promoting the uptake of AI in the transport sector ~~and to propose, if necessary in order to uphold EU fundamental rights,~~, to ensure that EU fundamental rights are upheld and to propose corresponding changes to EU legislation without delay and in close cooperation with all stakeholders in the transport sector;

Amendment 39 #

Draft opinion
Paragraph 4 b (new)

4b. Notes that data security and privacy will come along with ethical concerns regarding the definition of the data to collect as well as their ownership, sharing, storage and purpose; notes, additionally, that ethics will play a key role in the definition of the legislative framework regulating the use and management of such data;

Amendment 4 #

Draft opinion
Recital B

B. whereas a European approach to ~~AI needs to include ethical aspects of~~artificial intelligence should address, as a matter of priority, the ethical aspects and dilemmas associated with AI to ensure that it is human-centric, enhances human well- being, a sense of security, the well-being of society and the environment, and fully respects EU fundamental rights and values;

Amendment 4 #

Draft opinion
Citation 1 d (new)

- Whereas transparent, accountable and inclusive processes to draft, enact and evaluate policies and legislation applicable to the design, development and deployment of algorithmic systems are of utmost importance to ensure that all individuals that are directly impacted have a meaningful say in whether and how these systems are used, and in whose interest;

Amendment 40 #

Draft opinion
Paragraph 3 a (new)

3 a. Stresses that the development and spread of AI must not restrict access to transport services; points out that the cost of this technology will create a significant barrier to the ownership of autonomous cars and will create the need for a public system offering an access service to such vehicles in order to combat traffic exclusion;

Amendment 40 #

Draft opinion
Paragraph 4 c (new)

4c. Reiterates European principles on the ownership of individuals of their own personal data and explicit, informed consent which is necessary before using personal data as enshrined in the GDPR; points out that consent implies that individuals understand for which purpose their data will be used and that entities using personal data in algorithms have a responsibility for ensuring this understanding;

Amendment 41 #

Draft opinion
Paragraph 3 a (new)

3 a. Stresses that AI systems should not damage nor hurt human physical and psychological integrity; calls therefore on AI systems to be technically robust in order for them not to be used for harmful purposes;

Amendment 41 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 5

5. Believes that consumers ~~should be adequately informed in a timely, impartial, easily-readable, standardised~~have the right to be informed in an understandable, and accessible manner about the existence, process, ~~rationale,~~and reasoning ~~and possible outcome~~ of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected; believes that consumers should also be protected by the right to switch off or limit an AI system using personalisation where possible;

Amendment 42 #

Draft opinion
Paragraph 4

4. Recommends the development of an EU-wide ~~trustworthy AI label for the automotive industry, which should provide for common and harmonised standards~~ethical standards for all modes of transport, including the automotive industry, on safety, technical robustness, privacy and transparency and for testing of AI-enabled vehicles and related products and services;

Amendment 42 #

Maria da Graça CARVALHO, Edina TÓTH

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence~~, process, rationale, reasoning and possible outcome~~ and reasoning of algorithmic systems, ~~about~~on how to reach a human with decision- making powers, and about how the system’s decisions can be checked, ~~meaningfully~~ contested and corrected;

Amendment 43 #

Draft opinion
Paragraph 4

4. Recommends the development of an EU-wide trustworthy AI ~~label~~standards for the automotive industry, which should provide for common and harmonised standards on safety, interoperability, technical robustness - including concerns of resource efficient use, reparability and recyclability, privacy and transparency and for testing of AI-enabled vehicles and related products and services;

Amendment 43 #

Draft opinion
Paragraph 5

5. BUnderlines that consumer trust is essential for the development and implementation of artificial intelligence, robotics and related technologies which can carry inherent risks when they are based on opaque algorithms and biased data sets; believes that consumers should be adequately informed in a timely, ~~impartial~~accurate, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;

Amendment 44 #

Draft opinion
Paragraph 4

4. Recommends the development of an EU-wide trustworthy AI label for the automotive industry, which should provide for common and harmonised standards on safety, technical robustness, privacy, data protection and transparency and for testing of AI-enabled vehicles and related products and services;

Amendment 44 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected; stresses the importance of proportionality in the development of a transparency framework to avoid unnecessary burdens on SMEs operating in low-risk sectors and to protect intellectual property rights;

Amendment 45 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Marc ANGEL, Alex AGIUS SALIBA, Adriana MALDONADO LÓPEZ, Brando BENIFEI

Draft opinion
Paragraph 4 a (new)

4 a. Points out that the development of artificial intelligence technologies in the transport sector raises serious ethical challenges for the future of the labour market, particularly in the area of passenger and goods transport; recommends that the Commission take this problem into account in its impact assessments of AI legislation;

Amendment 45 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and ~~possible~~intended and reasonably foreseeable unintended outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected; this information should not, however, endanger trade secrets and other competitive advantages;

Amendment 46 #

Draft opinion
Paragraph 4 a (new)

4 a. Considers that AI should be sustainably respectful with the environment as well as it should promote research to achieve UN’s Sustainable Development Objectives;

Amendment 46 #

Christel SCHALDEMOSE, Leszek MILLER, Evelyne GEBHARDT, Marc ANGEL, Alex AGIUS SALIBA, Sándor RÓNAI, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Andreas SCHIEDER, Sylvie GUILLAUME, Brando BENIFEI

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected; recalls that humans must always be able to overrule automated decisions that are final and permanent;

Amendment 47 #

Evelyne GEBHARDT, Maria-Manuel LEITÃO-MARQUES, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Sylvie GUILLAUME, Sándor RÓNAI

Draft opinion
Paragraph 5

5. Calls on the Commission to explore the possibility of entrusting a relevant existing EU agency with enforcement and sanction mechanisms, so that actions can be taken if an AI system used in transport violates fundamental rights or the European ethical and security framework;deleted

Amendment 47 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, the aims or purpose, process, rationale, reasoning and possible outcome and consequences for consumers of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;

Amendment 48 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 5

5. Calls on the Commission to explore ~~the possibility of entrusting a relevant existing EU agency with enforcement and sanction mechanisms, so that actions can be take~~how the existing instruments of supervision and control in the transport sector can be equipped and used to take action if an AI system used in transport violates fundamental rights or the European ethical and security framework;

Amendment 48 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed, if requested in offline format, in a timely, impartial, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;

Amendment 49 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ

Draft opinion
Paragraph 5

5. Calls on the Commission to explore the possibility of e~~ntrusting a relevant existing EU agency with enforcement and sanction mechanisms, so that actions can be take~~stablishing a EU body, made up of the various member state regulators to ensure oversight on European level in order to enable Commission to take action if an AI system used in transport violates fundamental rights or the European ethical and security framework;

Amendment 49 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence~~, process, rationale, reasoning and possible outcome~~ of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;

Amendment 5 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Evelyne GEBHARDT, Marc ANGEL, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Sándor RÓNAI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Recital B

B. whereas a European approach to AI needs to include ethical aspects of AI to ensure that it is human-centric, enhances human well-being, the well-being of society and the environment, and fully respects EU fundamental rights and values; whereas the protection of the right to privacy and data protection should be mandatory and that data collection should be fully in line with the GDPR and the ePrivacy regulation, stored for a limited time, and be effectively anonymised;

Amendment 5 #

Draft opinion
Paragraph 1

1. Underlines the importance of an EU regulatory framework being applicable where consumers within the Union are users of or subject to an algorithmic system, irrespective of the place of establishment of the entities that develop, sell or employ the system; furthermore, believes that the rules set out should apply across the value chain, namely development, deployment and use of the relevant technologies and their components and for all developers, and should guarantee the highest level of consumer protection; proposes that these rules take into account the lessons drawn from the adaptation to Regulation (EU) 2016/6791a (GDPR), considered a global benchmark; considers the designation of a liable entity in the Union (such as authorised representative) important for its enforcement; __________________ 1aRegulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).

Amendment 50 #

Draft opinion
Paragraph 5

5. Calls on the Commission to explore the possibility of entrusting a relevant existing EU agency with monitoring enforcement ~~and sanction mechanism~~mechanisms and imposing sanctions, so that actions can be taken if an AI system used in transport violates fundamental rights or the European ethical and security framework;

Amendment 50 #

Draft opinion
Paragraph 5

5. Believes that consumers should be adequately informed in a timely, ~~impartial~~accurate, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;

Amendment 51 #

Draft opinion
Paragraph 5 a (new)

5 a. Calls on the Commission to work closely with Member States on the design, implementation and enforcement of European ethical and safety standards for AI; notes that the EU has the potential to become a global leader in promoting a socially responsible approach to this technology and its use;

Amendment 51 #

Draft opinion
Paragraph 5 a (new)

5a. Stresses the need to effectively address the challenges created by algorithmic systems and to ensure that consumers are empowered and properly protected; underlines the need to look beyond the traditional principles of information and disclosure on which the consumer acquis has been built, as stronger consumer rights and clear limitations regarding the development and use of algorithmic systems will be necessary to ensure technology contributes to making consumers’ lives better and evolves in a way that respects fundamental and consumer rights and European values;

Amendment 52 #

Draft opinion
Paragraph 6

6. Calls on the Commission to further support the development of trustworthy AI systems in order to render transport more efficient, accessible, affordable and inclusive, including for persons with reduced mobility and other disabilities, taking particular account of Directive (EU) 2019/882 on the accessibility requirements for products and services and of European legislation on passenger rights.

Amendment 52 #

Draft opinion
Paragraph 5 a (new)

5a. Highlights the need to pay particular attention to situations involving more vulnerable groups such as children, persons with disabilities, elderly people and others that have historically been disadvantaged or are at risk of exclusion, and to situations which are characterised by asymmetries of power or information, such as between employers and workers, or between businesses and consumers;

Amendment 53 #

Maria da Graça CARVALHO, Edina TÓTH, Róża THUN UND HOHENSTEIN

Draft opinion
Paragraph 6

Amendment 53 #

Draft opinion
Paragraph 5 a (new)

5a. Considers that a value-sensitive design approach is strongly needed to allow a widespread social acceptance of AI for consumers; considers that ethical values of fairness, accuracy, confidentiality and transparency should be the basis of AI which in this context entails that the system’s operations cannot generate unfairly biased outputs;

Amendment 54 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 6

6. Calls on the Commission to further support the development of trustworthy AI systems in order to render transport safer, more efficient, accessible, affordable and inclusive, including for persons with reduced mobility.

Amendment 54 #

Evelyne GEBHARDT, Maria-Manuel LEITÃO-MARQUES, Marc ANGEL, Monika BEŇOVÁ

Draft opinion
Paragraph 5 a (new)

5a. Believes that a European legal framework for a system of risk classification and related legal obligations is required to ensure a uniform protection of European consumers;

Amendment 55 #

Draft opinion
Paragraph 6 a (new)

6 a. Notes that the development and implementation of AI in the transport sector will not be possible without modern infrastructure, which is an essential part of intelligent transport systems; stresses that the persistent divergences in the level of development between Member States create the risk of depriving the least developed regions and their inhabitants of the benefits brought by the development of autonomous mobility; calls for the modernisation of infrastructure in the EU, including its integration into the 5G network, to be adequately funded;

Amendment 55 #

Draft opinion
Paragraph 5 b (new)

5b. Believes that explicability is crucial for building and maintaining users’ trust in AI systems; this calls for processes to be transparent, the capabilities and purpose of AI systems openly communicated, and decisions explainable to those directly and indirectly affected; believes that ‘black box’ algorithms which do not provide such information must be required to provide that the system as a whole respects fundamental rights;

Amendment 56 #

Draft opinion
Paragraph 6 a (new)

6 a. notes that at sectors like public transportation, AI systems for intelligent transport systems can be used to minimise queuing, optimise routing, allow vision impaired people to be more independent, optimise energy efficient engines and thereby enhance decarbonisation efforts and reduce the environmental footprint, for a greener society;

Amendment 56 #

Draft opinion
Paragraph 6

6. Recalls the importance of ensuring the availability of effective remedies for consumers and calls on the Member States to ensure that accessible, affordable, independent and effective procedures are available to guarantee an impartial review of all claims of violations of consumer rights through the use of algorithmic systems, whether stemming from public or private sector actors; urges that dispute resolution and collective redress mechanisms should be made available both offline and online to consumers, groups and legal entities, who wish to contest the introduction or ongoing use of a system with potential for consumer rights violations, or remedy a violation of rights;

Amendment 57 #

Draft opinion
Paragraph 6 a (new)

6 a. Underlines the critical importance for data science in order to design discrimination free AI systems and prevent tainted data to be used;

Amendment 57 #

Draft opinion
Paragraph 6

6. Recalls the importance of ensuring the availability of effective remedies for consumers and calls on the Member States to ensure that accessible, affordable, independent and effective procedures are available to guarantee an impartial review of all claims of violations of consumer rights through the use of algorithmic systems, whether stemming from public or private sector actors; urges Member States to ensure consumer organisations have sufficient funding to assist consumers to exercise their right to remedy;

Amendment 58 #

Maria da Graça CARVALHO, Edina TÓTH, Róża THUN UND HOHENSTEIN, Andrey KOVATCHEV, Marion WALSMANN, Pascal ARIMONT

Draft opinion
Paragraph 6 b (new)

6 b. Notes that currently, worldwide, one human dies every 23 seconds in a car accident. AI systems could help to reduce the number fatalities significantly, for instance through better reaction times and better adherence to rules. In order to maximalise the potential of AI in transport sector clear ethical guidelines are absolutely essential.

Amendment 58 #

Draft opinion
Paragraph 6

6. Recalls the importance of ensuring the availability of effective remedies for consumers and calls on the Member States and national market surveillance authorities to ensure that accessible, affordable, independent and effective procedures and review structures are available to guarantee an impartial human review of all claims of violations of consumer rights through the use of algorithmic systems, whether stemming from public or private sector actors;

Amendment 59 #

Virginie JORON, Jean-Lin LACAPELLE, Alessandra BASSO, Marco CAMPOMENOSI

Draft opinion
Paragraph 6 b (new)

6 b. Recommends to follow procedures for data processing compliant with existing legislation, confidentiality, anonymity, fair treatment and data protection regulation;

Amendment 59 #

Draft opinion
Paragraph 7

7. Stresses that where public money contributes to the development or implementation of an algorithmic system, the code, the generated data -as far as it is non-personal- and the trained model should be public by default, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;deleted

Amendment 6 #

Tom BERENDSEN, Marian-Jean MARINESCU, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Jens GIESEKE, Massimiliano SALINI, Jörgen WARBORN

Draft opinion
Recital B

B. whereas a European approach to AI ~~needs to include ethical aspects of AI to ensure that it is~~, robotics and related technologies needs to be in accordance with ethical principles reflecting our European values and principles to ensure that AI, robotics and related technologies are human- centric, enhances human well-being, the well-being of society and the environment, and fully respects EU fundamental rights and values;

Amendment 6 #

Maria da Graça CARVALHO, Edina TÓTH, Andrey KOVATCHEV, Marion WALSMANN, Pascal ARIMONT

Draft opinion
Paragraph 1

1. Underlines the importance of an EU regulatory framework of ethical aspects of artificial intelligence , robotics and related technologies being applicable where consumers within the Union are users of or subject to an algorithmic system, irrespective of the place of establishment of the entities that develop, sell or employ the system in order to bring legal certainty to business and citizens alike;

Amendment 60 #

Draft opinion
Paragraph 6 c (new)

6 c. Draws attention to the high added value provided by autonomous vehicles for persons with reduced mobility, as such vehicles allow them to participate more effectively in individual road transport and thereby facilitate their daily lives;

Amendment 60 #

Draft opinion
Paragraph 7

Amendment 61 #

Draft opinion
Paragraph 6 d (new)

6 d. Takes into account that it will be impossible for autonomous vehicles to eliminate all accidents; this makes explicability increasingly important in order to justify results, particularly when unexpected decisions are made;

Amendment 61 #

Draft opinion
Paragraph 7

Amendment 62 #

Draft opinion
Paragraph 6 e (new)

6 e. Takes the view that it must always be possible to explain the AI decisions as well as any data driving those decisions to end-users and other stakeholders in non- technical terms;

Amendment 62 #

Draft opinion
Paragraph 7

7. Stresses that where public money contributes to the development or implementation of an algorithmic system, alongside open procurement and open contracting standards, the code, the generated data -as far as it is non-personal- and the trained model should be public by default, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;

Amendment 63 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Marc ANGEL, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 6 f (new)

6 f. Stresses the importance of accessibility, especially when designing MaaS-systems (Mobility as a Service).

Amendment 63 #

Draft opinion
Paragraph 7

7. Stresses that where ~~public money contributes~~money originating from public sources contributes significantly to the development or implementation of an algorithmic system, the code, the generated data - as far as it is non-personal - and the trained model should be public by default, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;

Amendment 64 #

Maria da Graça CARVALHO, Edina TÓTH, Andrey KOVATCHEV

Draft opinion
Paragraph 7

7. Stresses that where public money contributes to the development or implementation of an algorithmic system, the code, the generated data -as far as it is non-personal- and the trained model should be public by default, to ~~enabl~~guarantee transparency and ~~reuse, among other goals, to maximise the achievement~~enable their reuse to foster innovation; stresses that in this way, the full potential of the Single Market~~, and to~~ can be unlocked avoiding market fragmentation;

Amendment 65 #

Draft opinion
Paragraph 7

7. Stresses that where public money contributes to the development or implementation of an algorithmic system, the code, the generated data -as far as it is non-personal- and the trained model should be ~~public by default~~accessible, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;

Amendment 66 #

Maria da Graça CARVALHO, Róża THUN UND HOHENSTEIN, Edina TÓTH, Andrey KOVATCHEV, Marion WALSMANN, Pascal ARIMONT

Draft opinion
Paragraph 7

7. Stresses that where public money contributes to the development or implementation of an algorithmic system, ~~the code,~~ the generated data -as far as it is non- personal- and the trained model should be public by default, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;

Amendment 67 #

Draft opinion
Paragraph 7 a (new)

7a. Recalls that an examination of the current EU legal framework, including the consumer law acquis, data protection legislation, product liability legislation, product safety and market surveillance legislation, is needed to check that it is able to respond to the emergence of AI and automated decision-making and that it is able to provide a high level of consumer protection;

Amendment 68 #

Draft opinion
Paragraph 7 a (new)

7a. Stresses that the data sets and the processes that yield the AI system’s decision, including those of data gathering and data labelling as well as the algorithms used, should be documented to the best possible standard to allow for traceability and an increase in transparency; stresses that this also applies to the decisions made by the AI system;

Amendment 69 #

Christel SCHALDEMOSE, Leszek MILLER, Evelyne GEBHARDT, Marc ANGEL, Alex AGIUS SALIBA, Sándor RÓNAI, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Andreas SCHIEDER, Sylvie GUILLAUME, Brando BENIFEI

Draft opinion
Paragraph 7 a (new)

7a. Underlines that the increased use of artificial intelligence requires a strong focus on digital security, as the large amount of data creates new risks of cyberattacks; calls on the Commission to develop clear guidelines for businesses and public agencies to take the necessary precautions when using artificial intelligence;

Amendment 7 #

Draft opinion
Recital C a (new)

C a. whereas the full potential of AI in the transport sector can only be exploited if users are aware of the potential benefits and challenges that this technology brings; whereas it is necessary to incorporate this issue into the education process, including in terms of combating digital exclusion, and to conduct information campaigns at European level that give an accurate representation of all aspects of AI development;

Amendment 7 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Jordi CAÑAS, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 1

1. Underlines the importance of an EU regulatory framework of ethical aspects being applicable where consumers within the Union are users of or subject to an algorithmic system, irrespective of the place of establishment of the entities that develop, sell or employ the system;

Amendment 70 #

Evelyne GEBHARDT, Maria-Manuel LEITÃO-MARQUES, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME, Sándor RÓNAI

Draft opinion
Paragraph 7 a (new)

7a. Asks the Commission to issue binding rules for companies to publish transparency reports including the existence, functionality, process, main criteria, the logic behind, the datasets used and possible outcome of algorithmic systems and efforts to identify, prevent and mitigate discrimination in algorithmic systems in a timely, impartial, easily-readable, and accessible manner;

Amendment 71 #

Draft opinion
Paragraph 7 a (new)

7a. Calls for a uniform implementation of the system of risk classification and related legal obligations to ensure a level-playing field among the Member States and to prevent a fragmentation of the internal market;

Amendment 72 #

Virginie JORON, Jean-Lin LACAPELLE, Marco CAMPOMENOSI

Draft opinion
Paragraph 7 a (new)

7a. Stresses that where European public funds contribute to the development or implementation of an algorithmic system of a consortium comprising a non-European company, the code and the non-personal data generated should be public by default.

Amendment 73 #

Virginie JORON, Jean-Lin LACAPELLE, Marco CAMPOMENOSI

Draft opinion
Paragraph 7 b (new)

7b. Underlines that data sets used by AI systems (both for training and operation) may suffer from the inclusion of inadvertent historic bias, incompleteness and bad governance models; stresses that the continuation of such biases could lead to unintended (in)direct prejudice and discrimination against certain groups or people, potentially exacerbating prejudice and marginalisation; notes that harm can also result from the intentional exploitation of (consumer) biases or by engaging in unfair competition, such as the homogenisation of prices by means of collusion or a non-transparent market; stresses that identifiable and discriminatory bias should be removed in the collection phase where possible; notes that the way in which AI systems are developed (e.g. algorithms’ programming) may also suffer from unfair bias; stresses that this could be counteracted by putting in place oversight processes to analyse and address the system’s purpose, constraints, requirements and decisions in a clear and transparent manner; notes that hiring from diverse backgrounds should be encouraged;

Amendment 74 #

Draft opinion
Paragraph 7 b (new)

7b. Stresses that where non-European public funds contribute to the development or implementation in Europe of an algorithmic system, the code and the non-personal data generated should be public by default.

Amendment 75 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Evelyne GEBHARDT, Marc ANGEL, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Sándor RÓNAI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 7 b (new)

7b. Considers that AI, IoT, and other emerging technologies have enormous potential to deliver opportunities for consumers to have access to several amenities in many aspects of their lives alongside with better products and services, as well as to benefit from better market surveillance, as long as all applicable principles, conditions (including transparency and auditability), and regulations continue to apply;

Amendment 76 #

Evelyne GEBHARDT, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Maria GRAPINI, Sylvie GUILLAUME, Sándor RÓNAI

Draft opinion
Paragraph 7 b (new)

7b. Calls for the establishment of an Union-wide registration system for artificial intelligence, robotics and related technologies to support the uniform and transparent implementation of the risk classification in the Union;

Amendment 77 #

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy lawall consumers and groups, including marginalised and vulnerable, are adequately taken into account; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 78 #

Jordi CAÑAS, Svenja HAHN

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups, such as persons with disabilities, are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 79 #

Tom BERENDSEN, Marian-Jean MARINESCU, Barbara THALER, Sven SCHULZE, Andor DELI, Elżbieta Katarzyna ŁUKACIJEWSKA, Massimiliano SALINI, Jens GIESEKE, Jörgen WARBORN

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection, intellectual property rights and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 8 #

Draft opinion
Recital D

D. whereas the EU aimed to reduce annual road fatalities in the EU by 50% by 2020 compared to 2010, but, in view of stagnating progress, renewed its efforts in its Road Safety Policy Framework 2021 - 2030 - Next steps towards "Vision Zero"; whereas in this regard, AI, automation and other new technologies have a great potential to increase road safety by ~~avoiding~~potentially reducing the possibilities for human error;

Amendment 8 #

Draft opinion
Paragraph 1

1. Underlines the importance of an EU regulatory framework being applicable where consumers within the Union are ~~users of or subject to~~directed towards or targeted by an algorithmic system, irrespective of the place of establishment of the entities that develop, sell or employ the system;

Amendment 80 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Sylvie GUILLAUME, Marc ANGEL, Alex AGIUS SALIBA, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Brando BENIFEI, Andreas SCHIEDER

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, where possible via APIs, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 81 #

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of consumers and groups which are marginalised ~~and~~or in vulnerable ~~consumers and group~~situations are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 82 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups are adequately taken into account and represented in any future regulatory framework of ethical aspects; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 83 #

Virginie JORON, Jean-Lin LACAPELLE, Marco CAMPOMENOSI

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and groups are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of ~~training and giving basic skills to consumers to~~educating consumers to be more informed when dealing with algorithmic systems in order to protect them from potential risks and ~~detriment of~~uphold their rights;

Amendment 84 #

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of marginalised and vulnerable consumers and ~~group~~European citizens are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably ~~independent researchers, media and civil~~university or socie~~ty organisation~~ntific researchers, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 85 #

Maria da Graça CARVALHO, Edina TÓTH, Marion WALSMANN, Pascal ARIMONT

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of ~~marginalised and vulnerable consumers and group~~consumers in vulnerable situations are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to non-personal data shcould be extended to ~~appropriate parties~~ notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 86 #

Maria-Manuel LEITÃO-MARQUES, Christel SCHALDEMOSE, Evelyne GEBHARDT, Marc ANGEL, Adriana MALDONADO LÓPEZ, Maria GRAPINI, Andreas SCHIEDER

Draft opinion
Paragraph 8

8. Underlines the importance of ensuring that the interests of ~~marginalised and vulnerable consumers and group~~consumers are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection ~~and~~, privacy law and business secrecy rules; recalls the importance of training and giving basic digital skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;

Amendment 87 #

Draft opinion
Paragraph 8 a (new)

8a. Considers that, when it comes to filling the gaps on the practical implementation of ethical guidelines on algorithmic systems and connected technologies, models such as the so-called VCIO model (Values, Criteria, Indicators, Observables) should be studied and evaluated in terms of fitness for purpose; further considers that such models to concretise and implement AI system requirements, as well as to produce labels that allow companies to communicate the ethical properties of their products clearly and uniformly through a standardised risk matrix, could enhance consumer literacy, information, and awareness;

Amendment 88 #

Maria da Graça CARVALHO, Andrey KOVATCHEV, Róża THUN UND HOHENSTEIN, Pascal ARIMONT

Draft opinion
Paragraph 8 a (new)

8a. Stresses the importance of ensuring that the collective rights of national and linguistic minorities are taken into account and represented in any future regulatory framework, in particular in relation to the regional and minority languages of the Union;

Amendment 89 #

Draft opinion
Paragraph 9

9. Underlines the importance of training highly skilled professionals in this area ~~and ensuring~~, the need of having diverse teams of developers and engineers working alongside with key actors to prevent gender and cultural bias of being inadvertently included in AI algorithms, systems and applications and ensure the mutual recognition of such qualifications across the Union;

Amendment 9 #

Draft opinion
Recital D

D. whereas the EU aimed to reduce annual road fatalities in the EU by 50% by 2020 compared to 2010, but, in view of stagnating progress, renewed its efforts in its Road Safety Policy Framework 2021 - 2030 - Next steps towards "Vision Zero"; whereas in this regard, automation and new technologies ~~have a great potential to increase~~are of vital importance in increasing the level of road safety by avoiding human error;

Amendment 9 #

Draft opinion
Paragraph 2

2. Notes that the framework should apply to algorithmic systems, including the fields of artificial intelligence (AI), internet of things (IoT), machine learning, deep learning, automated decision making processes and robotics; further notes that referral systems should be developed to help explain those systems to consumers whenever they present complexity or constitute decisions that impact their lives significantly;

Amendment 90 #

Draft opinion
Paragraph 9

9. Underlines the importance of training highly skilled professionals in this area and ensuring the mutual recognition of such qualifications across the Union; supports the creation of educational curricula and public awareness activities around the societal, legal, and ethical impact of AI;

Amendment 91 #

Svenja HAHN, Claudia GAMON, Dita CHARANZOVÁ, Jordi CAÑAS, Sandro GOZI, Stéphanie YON-COURTIN

Draft opinion
Paragraph 9

9. Underlines the importance of achieving a high-level of overall digital literacy and of training highly skilled professionals in this area ands well as ensuring the mutual recognition of such qualifications across the Union;

Amendment 92 #

Draft opinion
Paragraph 9

9. Underlines the importance of upskilling and training highly skilled professionals in this area and ensuring the mutual recognition of such qualifications across the Union;

Amendment 93 #

Draft opinion
Paragraph 9 a (new)

9a. Calls on the Commission to promote and fund the development of human-centric artificial intelligence, robotics and related technologies that address environment and climate challenges and that ensure equal access to and enjoyment of fundamental rights through the use of tax, procurement, or other incentives;

Amendment 94 #

Draft opinion
Paragraph 9 a (new)

9a. Notes that those who own or operate inputs to AI systems and profit from it should be asked to help fund the development of AI literacy programs for consumers as this is in the best interest of both the company and society as a whole;

Amendment 95 #

Draft opinion
Paragraph 9 b (new)

9b. Notes that particular attention in AI literacy programs must also be paid to situations where AI systems can cause or exacerbate adverse impacts due to asymmetries of power or information, such as between employers and employees, businesses and consumers or governments and citizens;

Amendment 96 #

Virginie JORON, Jean-Lin LACAPELLE, Marco CAMPOMENOSI, Alessandra BASSO

Draft opinion
Paragraph 9 b (new)

9b. Underlines that artificial intelligence and algorithmic systems should be legally compliant, robust, reliable and secure by design; calls on the Commission to ensure that the Union’s regulatory approach to algorithmic systems includes appropriate measures to enable that these systems are subject to independent control and oversight;

Amendment 97 #

Draft opinion
Paragraph 10

~~10. Calls for the Union to establish a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities;~~deleted

Amendment 98 #

Draft opinion
Paragraph 10

~~10. Calls for the Union to establish a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities;~~deleted

Amendment 99 #

Evelyne GEBHARDT, Marc ANGEL, Monika BEŇOVÁ, Alex AGIUS SALIBA, Andreas SCHIEDER, Sylvie GUILLAUME, Sándor RÓNAI, Brando BENIFEI

Draft opinion
Paragraph 10

10. Calls for the ~~Union to~~ establishment of a European market surveillance structure for algorithmic systems i~~ssuing guidance, opinions and expertise to Member States’ authorities;~~n the framework of a European Agency for Artificial Intelligence, which the Parliament proposed in its resolution of 16 February 20171a, issuing guidance, opinions and expertise to Member States’ authorities, monitoring the implementation of relevant EU legislation, addressing potential consumer protection issues, identifying standards for best practice, and, where appropriate, making recommendations for regulatory measures; __________________ 1aEuropean Parliament resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics (2015/2103(INL)) (OJ C 252, 18.7.2018, p. 239).

source: 650.440

2020/05/29 JURI 651 amendments...

Amendment 1 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Citation 2 a (new)

- having regard to the Interinstitutional Agreement of 13 April 2016 on Better Law-Making (OJ L 123, 12.5.2016) and the Better Regulation Guidelines,

Amendment 10 #

Motion for a resolution
Recital A

A. whereas artificial intelligence, robotics and related technologies with the potential to ~~directly impact all aspects of our societies, including basic social and economic principles and values,~~generate opportunities for business and citizens while directly impact all aspects of our societies are being developed very quickly;

Amendment 100 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 4

4. Underlines that transparency and explainability isare essential to ensur~~ing that~~e citizens’ trust in these technologies, even if the degree of explainability is relative to the complexity of the technologies, and that it should be complemented by auditability and traceability; considers that the respect of these principles is a precondition to guarantee accountability;

Amendment 101 #

Motion for a resolution
Paragraph 4

4. Underlines that explainability, i~~s essential to ensuring that~~nterpretability, auditability, traceability, and transparency, as well as access to technology, data and computing systems underlying such technologies contribute to citizens’ trust ~~these~~in technolog~~ies~~y, even if the degree of explainability ismay be relative to the complexity of the technologies~~, and that it should be complemented by auditability and traceability~~;

Amendment 102 #

Motion for a resolution
Paragraph 4

Amendment 103 #

Motion for a resolution
Paragraph 4

4. Underlines that traceability of AI systems should be ensured. In this connection, explainability is essential to ensuring that citizens trust these technologies, even if the degree of explainability is relative to the complexity of the technologies~~, and that it should be complemented by auditability and traceability~~;

Amendment 104 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 4

Amendment 105 #

Motion for a resolution
Paragraph 4 a (new)

4a. Considers that citizens should be informed when interacting with a system using artificial intelligence in particular to personalise a product or service to its users, whether they can switch off or restrain the personalisation; considers, furthermore, that transparency measures should be accompanied, as far as this is technically possible, by clear and understandable explanations of the data used, of the algorithm, of its purpose, of its outcomes, and of its potential dangers;

Amendment 106 #

Motion for a resolution
Paragraph 4 a (new)

4a. Asks the Commission to issue binding rules for companies to publish transparency reports including the existence, functionality, process, main criteria, the logic behind the datasets used and possible outcome of algorithmic systems and efforts to identify, prevent and mitigate discrimination in artificial intelligence and automated decision making systems in a timely, impartial, easily-readable, and accessible manner;

Amendment 107 #

Motion for a resolution
Paragraph 4 b (new)

4b. Stresses that where public money essentially contributes to the development or implementation of artificial intelligence, robotics and related technologies, the code, the generated data - as far as it is non-personal - and the trained model should be public by default, in addition to open procurement and open contracting standards to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;

Amendment 108 #

Motion for a resolution
Subheading 4 a (new)

Redress

Amendment 109 #

Motion for a resolution
Paragraph 4 c (new)

4c. Recalls the importance of ensuring the availability of effective remedies for individuals and calls on the Member States to ensure that accessible, affordable, independent and effective procedures are available to guarantee an impartial review of all claims of violations of consumer, civil and equal rights through the use of algorithmic systems, whether stemming from public or private sector actors;

Amendment 11 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital A a (new)

Aa. whereas artificial intelligence, robotics and related technologies will also lead to substantial changes to the labour market, posing a serious challenge to the public authorities in terms of reorganisation of the workforce;

Amendment 110 #

Motion for a resolution
Paragraph 5

5. ~~Recalls that~~Is concerned by the risks of biases and discrimination in the development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies,; recalls that, in all circumstances, they should respect human dignity and ensure equal treatment for all;

Amendment 111 #

Motion for a resolution
Paragraph 5

5. Recalls that the development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by ~~such technologies, should respect human dignity~~them, should respect Union law, as well as human rights and dignity, autonomy and self-determination of the individual, and ensure equal treatment and non- discrimination for all;

Amendment 112 #

Motion for a resolution
Paragraph 5

5. Recalls that the development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should respect human dignity and ensure equal treatment for all;

Amendment 113 #

Motion for a resolution
Paragraph 5 a (new)

5a. Stresses that AI systems should be designed to respect, serve and protect moral values and physical and mental integrity, promote a sense of personal and cultural identity and help satisfy essential needs; underlines the need to avoid any use of AI systems that might lead to inadmissible direct or indirect coercion, threaten to undermine psychological autonomy and mental health or lead to unjustified surveillance, deception or inadmissible manipulation;

Amendment 114 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 5 a (new)

5a. Firmly believes that there is an urgent need to examine how fundamental human rights texts and principles, as well as the resultant obligations, should be applied effectively to ensure that these emerging technologies do not create gaps in terms of protection;

Amendment 115 #

Motion for a resolution
Paragraph 6

6. Affirms that possible bias in and discrimination by software, algorithms and data should be addressed by setting rules for the processes through which they are designed and used~~, as this approach would have the potential to turn software, algorithms and data into a considerable counterbalance to bias and discrimination, and a positive force for social change~~;

Amendment 116 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 6

6. Affirms that possible bias in and discrimination by software, algorithms and data ~~should be addressed by setting~~can cause manifest harm to individuals and to society, therefore they should be addressed by encouraging the development and sharing of strategies to counter these, such as de-biasing datasets in research and development, and by the development of rules for the processes through which they are designed and used~~, a~~; considers this approach ~~would~~to have the potential to turn software, algorithms and data into a ~~considerable counterbalance to~~n asset in combating bias and discrimination, and a ~~positive~~ force for positive social change;

Amendment 117 #

Motion for a resolution
Paragraph 6

6. ~~Affirm~~Considers that possible bias in and discrimination by software, and algorithms ~~and data sh~~could be addressed by setting rules ~~for the~~on data processes through which they are designed and used, as this approach would have the potential to turn software, algorithms and data into a considerable counterbalance to bias and discrimination, and a positive force for social changeing; points out also that the use of AI, robotics and related technologies has the potential to fight discrimination in certain situation;

Amendment 118 #

Motion for a resolution
Paragraph 6

6. Affirms that possible bias in and discrimination by software, algorithms and data should be addressed by setting rules for the processes through which they are designed and used, as this approach would have the potential to turn software, algorithms and data into a considerable counterbalance to bias and discrimination, and a positive force for equal rights and social change;

Amendment 119 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 6

6. Affirms that possible bias in and discrimination by software, algorithms and data should be addressed by setting rules for the processes through which they are designed and used, as this approach would have the potential to turn software, algorithms and data into a considerable counterbalance to bias and discrimination~~, and a positive force for social change~~;

Amendment 12 #

Motion for a resolution
Recital A a (new)

Aa. whereas artificial intelligence, robotics and related technologies can make a huge contribution to reaching our common goal of improving the lives of citizens and fostering prosperity within the EU;

Amendment 120 #

Motion for a resolution
Paragraph 6

6. Affirms that possible bias in and discrimination by software, algorithms and data should be addressed by setting rules for the processes through which they are designed and used, as this approach would have the potential to turn software, algorithms and data into a considerable counterbalance to unfair bias and discrimination, and a positive force for social change;

Amendment 121 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 6 a (new)

6a. Underlines the importance of an ethical and regulatory framework including in particular provisions on the quality of data sets used for artificial intelligence and automated decision making systems depending on their context, especially regarding the representativeness of the training data, on the de-biasing of data sets, on the algorithms used, and on data and aggregation standards;

Amendment 122 #

Motion for a resolution
Subheading 5 a (new)

Right to redress

Amendment 123 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 6 a (new)

6a. Considers that any natural or legal person should be able to seek redress of a decision issued by a high-risk artificial intelligence, robotics or related technology at his or her detriment;

Amendment 124 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should safeguard and promote fundamental values of our society such as democracy, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativitysocial equality, labour rights and prosperity as well as fundamental rights and values;

Amendment 125 #

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should safeguard and promote fundamental values of our society such as democracy, freedom of expression and political pluralism, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity;

Amendment 126 #

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies ~~should~~has a role to play in contributing to find solutions that safeguard and promote fundamental values of our society such as democracy, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity;

Amendment 127 #

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should safeguard and promote fundamental values of our society such as democracy, the rule of law, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity, etc.;

Amendment 128 #

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should safeguard and promote fundamental rights and values of our society such as democracy, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, accessibility, digital literacy, innovation and creativity;

Amendment 129 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should safeguard and promote fundamental values of our society such as democracy, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, protection of children, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity;

Amendment 13 #

Motion for a resolution
Recital A b (new)

A. whereas, in areas such as health, agriculture, energy, transport, climate and various industrial processes artificial intelligence, robotics and related technologies can contribute to the development of better strategies and innovations;

Amendment 130 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 7

7. Emphasises that socially responsible artificial intelligence, robotics and related technologies should aim at safeguarding and promoteing fundamental values of our society such as democracy, diverse and independent media and objective and freely available information, health and economic prosperity, equality of opportunity, workers’ and social rights, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity;

Amendment 131 #

Motion for a resolution
Paragraph 7 a (new)

7a. Stresses the vital importance of guaranteeing freedom of thought and expression, to prevent some ideas that are not considered ‘politically correct’ from being deleted in advance through the use of certain software and algorithms, with the aim of promoting certain ideas and values to the detriment of others;

Amendment 132 #

Motion for a resolution
Paragraph 7 a (new)

7a. Stresses that the benefits of developing and using artificial intelligence, robotics and related technologies must be shared in such a way as to reduce social inequalities;

Amendment 133 #

Motion for a resolution
Paragraph 7 b (new)

7b. Asserts that the European model for developing artificial intelligence, robotics and related technologies must be based on citizens’ trust and greater social cohesion;

Amendment 134 #

Motion for a resolution
Paragraph 7 c (new)

7c. Objects to the deployment of artificial intelligence systems such as facial, biometric or emotion recognition in public spaces, with the exception of certain establishments which might be visited by the public and which, by their nature, pose a high security risk to individuals and the state;

Amendment 135 #

Motion for a resolution
Paragraph 7 d (new)

7d. Considers any artificial intelligence mechanism which hinders or restricts freedom of expression exercised digitally, except in cases of abuse of that freedom as laid down by law and sanctioned by the judiciary, to be illegal under the fundamental principles of the European Union;

Amendment 136 #

Motion for a resolution
Paragraph 7 e (new)

7e. Calls for a ban on the deployment of any artificial intelligence system which restricts users’ access to public services;

Amendment 137 #

Motion for a resolution
Paragraph 8

8. ~~Proposes that the~~Stresses the importance of responsible research and development aiming at maximizing a full potential of artificial intelligence, robotics and related technologies ~~in this regard should be maxim~~for citizedns and ~~explored through responsible research and innovation that requires the~~public good; calls for mobilisation of resources by the Union and its Member States in order to develop and support responsible innovation;

Amendment 138 #

Motion for a resolution
Paragraph 8 a (new)

8a. Stresses that technological expertise will be increasingly important and it will therefore be necessary continuously to update training courses, in particular for the new generations, and to promote the vocational retraining of people who are already working; innovation and training should be promoted not only in the private sector but also in the public sector;

Amendment 139 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 9

9. Insists that ~~the~~ develop~~ment~~er, deploy~~ment~~ers and users of these technologies should ~~not cause~~be held responsible for injury or harm of any kind to individuals or society in accordance with the relevant Union and national liability rules;

Amendment 14 #

Motion for a resolution
Recital A c (new)

Ac. whereas the development of artificial intelligence, robotics and related technologies is also a condition to reach the sustainability goals of the European Green Deal in many different sectors; whereas digital technologies can boost the impact of policies in delivering environmental protection;

Amendment 140 #

Motion for a resolution
Paragraph 9

9. Insists that the development, deployment and use of these technologies should not cause injury or harm of any kind to individuals or, society or environment;

Amendment 141 #

Motion for a resolution
Paragraph 9 a (new)

9a. Emphasises, in particular, that job losses resulting from the deployment of artificial intelligence systems must lead to a corresponding sharing of productivity gains and to appropriate public policies such as a reduction of working time;

Amendment 142 #

Motion for a resolution
Paragraph 10

10. States that it is essential that artificial intelligence, robotics and related technologies ~~support~~benefit the people and the planet, contribute to the achievement of sustainable development, the preservation of the environment, climate neutrality and circular economy goals; the development, deployment and use of these technologies should ~~be environmentally friendly, and contribute to minimising~~contribute to the green transition, preserve the environment, and minimise and remedy any harm caused to the environment during their lifecycle and across their entire supply chain;

Amendment 143 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 10

10. States that ~~it is essential that~~high-risk artificial intelligence, robotics and related technologies can contribute to finding solutions to support the achievement of sustainable development, climate neutrality and circular economy goals; the development, deployment and use of these technologies should be environmentally friendly, and contribute to minimising any harm caused to the environment during their lifecycle and across their entire supply chain in line with Union law;

Amendment 144 #

Motion for a resolution
Paragraph 10

10. States that ~~it is essential that~~ artificial intelligence, robotics and related technologies can be used by government and businesses to support the achievement of sustainable development, climate neutrality and circular economy goals; the development, deployment and use of these technologies should be environmentally friendly, and contribute to minimising any harm caused to the environment during their lifecycle and across their entire supply chain;

Amendment 145 #

Motion for a resolution
Paragraph 10 a (new)

10a. Considers that, for the purposes of the previous paragraph, the environmental impact of developing and deploying artificial intelligence, robotics and related technologies should always be evaluated throughout their lifetime and that the evaluation should include an estimate of the impact of the extraction of the materials needed, the energy consumption and the greenhouse gas emissions caused by their production, development and operation;

Amendment 146 #

Motion for a resolution
Paragraph 11

11. Proposes that for the purpose of developing responsible cutting-edge artificial intelligence solutions, the potential of artificial intelligence, robotics and related technologies ~~in this regard should be maximized and explor~~should be explored, stimulated and maximized through responsible research and innovation that requires the mobilisation of resources by the Union and its Member States;

Amendment 147 #

Motion for a resolution
Paragraph 11

11. Proposes that the potential of artificial intelligence, robotics and related technologies in this regard should be maximized and explored through responsible research and ~~innovation~~development that requires the mobilisation of resources by the Union and its Member States;

Amendment 148 #

Motion for a resolution
Paragraph 12

12. Highlights that the development, deployment and use of these technologies provide ~~numerous~~ opportunities for ~~the achievement~~promoting of the Sustainable Development Goals outlined by the United Nations, global energy transition and decarbonisation;

Amendment 149 #

Motion for a resolution
Paragraph 12

12. ~~Highlight~~Acknowledges that the development, deployment and use of these technologies provide ~~numerous~~ opportunities for the achievement of the Sustainable Development Goals outlined by the United Nations;

Amendment 15 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital B

B. whereas the Union and its Member States have a particular responsibility to make sure that these technologies contribute to the well-being and general interest of their citizens, while providing guarantees to enhance the added value of artificial intelligence in the EU;

Amendment 150 #

Motion for a resolution
Paragraph 12 a (new)

12a. Considers that the objectives of social responsibility, gender balance, environmental protection and sustainability should be without prejudice to existing general and sectorial obligations within these fields; believes that the Commission should establish non-binding guidelines to the intention of developers, deployers and users on the methodology for the achievement of these objectives;

Amendment 151 #

Motion for a resolution
Paragraph 12 a (new)

12a. Calls on the Commission to promote and fund the development of human-centric artificial intelligence, robotics and related technologies that address environment and climate challenges and that ensure the equal enjoyment of fundamental rights through the use of tax, procurement, or other incentives;

Amendment 152 #

Motion for a resolution
Paragraph 13

13. Observes that data production and use~~, including personal data such as biometric data,~~ resulting from the development, deployment and use of artificial intelligence, robotics and related technologies are rapidly increasing, thereby underlining the need to respect and enforce the rights of citizens to privacy and protection of personal data in line with Union law;

Amendment 153 #

Motion for a resolution
Paragraph 14

14. Points out that the possibility provided by these technologies of using personal ~~data~~ and non-personal data to categorise and micro-target people, identify ~~the~~ vulnerabilities of individuals, or exploit accurate predictive knowledge, has to be counterweighted by ~~the~~effectively enforced data protection and privacy principles ofsuch as data minimisation, the right to object to profiling and control one’s data, the right to obtain an explanation of a decision based on automated processing and privacy by design, as well as those of proportionality, necessity and limitation based on strictly identified purpose;

Amendment 154 #

Motion for a resolution
Paragraph 14

14. Points out that the possibility provided by ~~these~~high-risk technologies of using personal data and non-personal data to categorise and micro-target people, identify the vulnerabilities of individuals, or exploit accurate predictive knowledge, has to be counterweighted by the principles of data minimisation, the right to obtain an explanation of a decision based on automated processing and privacy by design, as well as those of proportionality, necessity and limitation based on purpose in compliance with GDPR;

Amendment 155 #

Motion for a resolution
Paragraph 15

Amendment 156 #

Motion for a resolution
Paragraph 15

15. Emphasises that when remote recognition technologies are used by public authorities during times of national emergency, such as during a national health crisis, their use should always be proportionate, limited in time and respectful of human dignity and, such as recognition of biometric features, notably facial recognition, are used by public authorities, their use should always be voluntary, proportionate, targeted and limited to specific objectives, restricted in time, and carried out in accordance with Union law, with due regard to human rights and dignity, autonomy and self- determination of the individual, and the fundamental rights set out in the Charter. Criteria for and limits to that use should be submitted to democratic scrutiny and debate, subjected to judicial review and respectful of human dignity, autonomy and self-determination of the individual, and further fundamental rights;

Amendment 157 #

Motion for a resolution
Paragraph 15

Amendment 158 #

Sergey LAGODINSKY, Alexandra GEESE, Patrick BREYER

Motion for a resolution
Paragraph 15

Amendment 159 #

Motion for a resolution
Paragraph 15 a (new)

15a. Calls on the Commission to propose a ban on artificial intelligence, automated decision making and related technologies for purposes of mass surveillance and for the use of biometric technologies by private actors in public spaces to safeguard the rights and freedoms guaranteed under the Charter of fundamental rights;

Amendment 16 #

Motion for a resolution
Recital B

Amendment 160 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Subheading 8 a (new)

Public power decisions

Amendment 161 #

Motion for a resolution
Paragraph 15 a (new)

15a. Points out that while the benefits of deploying artificial intelligence, robotics and related technologies within the framework of public power decisions are unquestionable, severe misuses are also possible, such as mass surveillance, predictive policing and breaches of due process rights;

Amendment 162 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 15 b (new)

15b. Considers that technologies which can replace decisions taken by public authorities should be treated with the utmost precaution, notably in the area of justice and law enforcement;

Amendment 163 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 15 c (new)

15c. Believes that Member States should have recourse to such technologies only if there is thorough evidence of their trustworthiness and if human verification is possible or systematic in cases where fundamental liberties are at stake; underlines the importance for national authorities to undertake strict fundamental rights impact assessment for high-risk artificial intelligence systems deployed in these cases;

Amendment 164 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER

Motion for a resolution
Paragraph 15 d (new)

15d. Is of the opinion that any decision taken by high-risk artificial intelligence, robotics or related technologies within the framework of prerogatives of public power should be subject to strict human verification and due process;

Amendment 165 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER

Motion for a resolution
Paragraph 15 e (new)

15e. Believes that the technological advancement should not allow for the use of artificial intelligence, robotics and related technologies to autonomously distribute rights or to impose legal obligations on individuals;

Amendment 166 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 16

16. Stresses that appropriate governance of the development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including by having measures in place focusing on accountability and addressing potential risks of unfair bias and discrimination, increases citizens’ safety and trust in those technologies;

Amendment 167 #

Motion for a resolution
Paragraph 16

16. Stresses that appropriate governance of the development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including by having measures in place focusing on accountability and addressing potential risks of bias and discrimination, increases citizens’ safety and trust in those technologies;

Amendment 168 #

Motion for a resolution
Paragraph 16

16. Stresses that appropriate governance of the development, deployment and use of artificial intelligence, robotics and related technologies, including by having measures in place focusing on accountability and addressing potential risks of bias and discrimination, can increases citizens’ safety and trust in those technologies;

Amendment 169 #

Motion for a resolution
Paragraph 16

16. Stresses that appropriate governance of the development, deployment and use of artificial intelligence, robotics and related technologies, including by having measures in place focusing on accountability and addressing potential risks ~~of bias and discrimination~~, increases citizens’ safety and trust in those technologies;

Amendment 17 #

Motion for a resolution
Recital B

B. whereas the Union and its Member States have a particular responsibility to make sure that these technologies are safe and contribute to the well-being and general interest of their citizens;

Amendment 170 #

Motion for a resolution
Paragraph 16 a (new)

16a. Considers that a common European framework for the governance of these technologies prepared by the European Commission or another relevant institution, body, office or agency of the Union (such as a European Agency for Artificial Intelligence, as called for below) designated for this task should be the basis for this governance in order to ensure a coherent European approach and prevent a fragmentation of the single market;

Amendment 171 #

Motion for a resolution
Paragraph 17

17. Observes that data are used in large volumes in the development of artificial intelligence, robotics and related technologies and that the processing, sharing of and access to such data must be governed in accordance with the ~~requirements of quality, integrity, security, privacy and control~~law;

Amendment 172 #

Motion for a resolution
Paragraph 17

17. Observes that data are used in large volumes in the development of artificial intelligence, robotics and related technologies and that the processing, sharing of ~~and~~, access to and use of such data must be governed in accordance with the requirements of quality, integrity, transparency, security, privacy and control;

Amendment 173 #

Motion for a resolution
Paragraph 17

17. Observes that data are used in large volumes in the development of artificial intelligence, robotics and related technologies and that the processing, sharing of and access to such data must be governed in accordance with the requirements of quality, integrity, interoperability, security, privacy and control;

Amendment 174 #

Motion for a resolution
Paragraph 17

17. Observes that data are used in large volumes in the development of high-risk artificial intelligence, robotics and related technologies and that the processing, sharing of and access to such data must be governed in accordance with the requirements of quality, integrity, security, privacy and control;

Amendment 175 #

Brando BENIFEI

Motion for a resolution
Paragraph 17 a (new)

17a. Recalls that access to data is an essential component in the growth of the digital economy; points out in this regard that interoperability of data, by limiting lock-in effects, plays a key role in ensuring fair market conditions and promoting a level playing field in the Digital Single Market;

Amendment 176 #

Esteban GONZÁLEZ PONS

Motion for a resolution
Paragraph 17 a (new)

17a. Points out in this regard that, if they are to be dependable, artificial intelligence, robotics and their related technologies must be technically robust and accurate;

Amendment 177 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 18

18. Underlines the need to ensure that data belonging to vulnerable groups, such as people with disabilities, patients, children, minorities ~~and migrants~~, migrants and other groups at risk of exclusion, are protected adequately;

Amendment 178 #

Motion for a resolution
Paragraph 18

18. Underlines the need to ensure that personal data is protected adequately, especially data belonging to vulnerable groups, such as people with disabilities, patients, children, minorities and migrants~~, are protected adequately~~;

Amendment 179 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 18

18. Underlines the need to ensure that data ~~belonging to~~on, or stemming from vulnerable groups, such as people with disabilities, patients, children, minorities and migrants, are ~~protected~~ adequately protected;

Amendment 18 #

Motion for a resolution
Recital B a (new)

Ba. whereas a European operational framework is of key importance in avoiding the fragmentation of the Single Market, resulting from differing national legislations; whereas, an action at European level will help fostering much needed investment, data infrastructure, research and common ethical norms; whereas this framework should be established according to the better regulation principle;

Amendment 180 #

Motion for a resolution
Paragraph 18 a (new)

18a. Highlights that such regulatory governance should also and specifically address the public sector, notably governments due to their unique position of power allowing them to employ algorithmic systems with major impact on citizens’ lives and fundamental rights;

Amendment 181 #

Motion for a resolution
Paragraph 18 a (new)

18a. Considers that the use of artificial intelligence, robotics and related technologies by public authorities requires specific attention because of its potential impact on fundamental rights protection;

Amendment 182 #

Motion for a resolution
Paragraph 18 b (new)

18b. Notes that the development, deployment and use of artificial intelligence, robotics and related technologies by public authorities are often outsourced to private parties; considers that this should not compromise the protection of public values and fundamental rights in any way; considers that public procurement terms and conditions should reflect the ethical standards imposed on public authorities;

Amendment 183 #

Motion for a resolution
Subheading 10

~~National~~Competent supervisory authorities

Amendment 184 #

Motion for a resolution
Paragraph 19

19. Notes the added value of having national supervisory authorities in each Member State responsible for ensuring, assessing and monitoring compliance with ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies; as well as for harmonising and balancing potential tensions between rights and ethics principles through methods of accountable deliberation;

Amendment 185 #

Motion for a resolution
Paragraph 19

19. Notes th~~e added value of having national~~at relevant or designated supervisory authorities in ~~each~~ Member States responsible for ensuring, assessing and monitoring compliance with legal and ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies can contribute to the compliance of development and research in artificial intelligence, robotics and related technologies;

Amendment 186 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 19

Amendment 187 #

Motion for a resolution
Paragraph 19

19. Notes the added value of having national supervisory ~~authoritie~~systems in each Member State responsible for ensuring, assessing and monitoring compliance with ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies;

Amendment 188 #

Motion for a resolution
Paragraph 19

19. Notes the ~~added value of having~~possible role of national supervisory authorities in each Member State responsible for ensuring, assessing and monitoring compliance with ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies;

Amendment 189 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 19 a (new)

19a. Notes that these authorities shall cooperate with the authorities responsible for implementing sectorial legislation in order to identify technologies which are high-risk from an ethical perspective and in order to supervise the implementation of required and appropriate measures where such technologies are identified;

Amendment 19 #

Motion for a resolution
Recital B b (new)

Bb. whereas such a framework should include legislative actions, where needed, including mandatory measures to prevent practices that would undoubtedly undermine fundamental rights and freedoms as defined in the Charter of Fundamental Rights of the European Union;

Amendment 190 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 19 a (new)

19a. Calls for such authorities to be tasked with promoting regular exchanges with civil society and innovation within the Union by providing assistance to concerned stakeholders, in particular small and medium-sized enterprises or start-ups;

Amendment 191 #

Esteban GONZÁLEZ PONS

Motion for a resolution
Paragraph 19 a (new)

19a. Points out that in all circumstances Member States must avoid duplicating the tasks of national authorities in so far as this is possible;

Amendment 192 #

Motion for a resolution
Paragraph 20

20. Indicates that such ~~authoritie~~national supervisory systems should liaise not only among themselves but also with the European Commission and other relevant institutions, bodies, offices and agencies of the Union in order to guarantee coherent cross-border action;

Amendment 193 #

Motion for a resolution
Paragraph 20

20. Indicates that such authorities ~~should~~may liaise not only among themselves but also with the European Commission and other relevant institutions, bodies, offices and agencies of the Union in order to guarantee coherent cross-border action;

Amendment 194 #

Motion for a resolution
Paragraph 20 a (new)

20a. Suggests that, in the context of such a cooperation, common criteria and an application process be developed for the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority;

Amendment 195 #

Motion for a resolution
Paragraph 21

Amendment 196 #

Motion for a resolution
Paragraph 21

21. Calls for such authorities to be tasked with promoting regular exchanges with civil society and innovation within the Union by providing assistance to concerned stakeholders, and less digitally- mature companies, in particular small and medium-sized enterprises or start-ups; notably towards awareness-raising and support for development, deployment, training and talent acquisition to ensure efficient technology transfer and access to technologies, projects, results and networks;

Amendment 197 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 21

21. Calls for such authorities to be tasked with promoting regular exchanges with civil society and innovation within the Union by providing assistance to relevant research and development, and concerned stakeholders, in particular small and medium-sized enterprises or start-ups;

Amendment 198 #

Motion for a resolution
Paragraph 21

21. Calls for such ~~authoritie~~national supervisory systems to be tasked with promoting regular exchanges with civil society and innovation within the Union by providing assistance to concerned stakeholders, in particular small and medium-sized enterprises or start-ups;

Amendment 199 #

Motion for a resolution
Paragraph 21

21. C~~alls for~~onsiders that such authorities tocould be tasked with promoting regular exchanges with civil society and innovation within the Union by providing assistance to concerned stakeholders, in particular small and medium-sized enterprises or start-ups;

Amendment 2 #

Motion for a resolution
Citation 2 a (new)

- having regard to Article 169 of the Treaty on the Functioning of the European Union,

Amendment 20 #

Motion for a resolution
Recital C

C. whereas a common framework for the development, deployment and use of artificial intelligence, robotics and related technologies within the Union should ~~both protect citizens from their potential risks and promote the trustworthiness of such technologies~~ allow citizens to share the benefits drawn from their potential, while protecting them, the public space and the environment from possible risks and promoting t~~he world~~rustworthy technology;

Amendment 200 #

Motion for a resolution
Subheading 11

~~A European Agency for Artificial Intelligence~~deleted

Amendment 201 #

Motion for a resolution
Subheading 11

~~A European Agency for Artificial Intelligence~~deleted

Amendment 202 #

Motion for a resolution
Subheading 11

~~A European Agency for Artificial Intelligence~~deleted

Amendment 203 #

Motion for a resolution
Subheading 11

~~A European Agency for Artificial Intelligence~~deleted

Amendment 204 #

Motion for a resolution
Paragraph 22

22. Recalls that Parliament’s resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics asked the Commission to consider the designation of a European Agency for Artificial Intelligence;deleted

Amendment 205 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 22

Amendment 206 #

Motion for a resolution
Paragraph 22

Amendment 207 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 22

Amendment 208 #

Motion for a resolution
Paragraph 22

Amendment 209 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph 22

Amendment 21 #

Motion for a resolution
Recital C

C. whereas a common European framework ~~for~~should ensure the development, the deployment and the use of ~~artificial intelligence, robotics and related technologies within the Union shoul~~trustworthy, ethical and technically robust artificial intelligence, based on Union’s laws and values and guided boy the pr~~otect citizens from their potential risks and promote the trustworthiness of such technologies in the world~~inciples of transparency and explainability, fairness, accountability and responsibility;

Amendment 210 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 23

Amendment 211 #

Motion for a resolution
Paragraph 23

Amendment 212 #

Motion for a resolution
Paragraph 23

Amendment 213 #

Motion for a resolution
Paragraph 23

Amendment 214 #

Motion for a resolution
Paragraph 23

Amendment 215 #

Motion for a resolution
Paragraph 23

Amendment 216 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 23

23. Calls on the Commission to follow- up on that request, especially in view of the added-value of having a body at Union level coordinating the mandates and actions of each national supervisory authority as referred to in the previous sub- section and drafting a common framework for the governance of the development, deployment and use of artificial intelligence, robotics and related technologies;

Amendment 217 #

Motion for a resolution
Paragraph 23

23. Calls on the Commission to ~~follow- up on that request, especially in view of the added-value of having a body at Union level coordinating the mandates and~~coordinate the actions of each national supervisory authority as referred to in the previous sub- section;

Amendment 218 #

Motion for a resolution
Paragraph 23

23. Calls on the Commission to follow- up on that request, especially in view of the added-value of having a body at Union level coordinating the ~~mandates and~~ actions, and sharing best practi~~ons~~ce of ~~each national~~the competent supervisory authorit~~y as referred to in the previous sub-section~~ies, and of contributing to the collaboration of research and development in the field throughout the EU;

Amendment 219 #

Motion for a resolution
Paragraph 23 a (new)

23a. Calls on the Commission to entrust an existing or new European Agency, or European body, with the task to ensure a harmonised implementation of the European ethical framework for high-risk artificial intelligence, robotics and related technologies;

Amendment 22 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital C a (new)

Ca. whereas, therefore, this common framework must make it possible to strike a balance with regard to the protection of citizens, maintaining confidence in the safety, reliability and consistency of goods and services, including digital technologies, and investment in innovative AI systems;

Amendment 220 #

Motion for a resolution
Paragraph 23 a (new)

23a. Calls on the Commission to assess whether a European body would be necessary to ensure a harmonised implementation of the European ethical framework for high-risk artificial intelligence, robotics and related technologies;

Amendment 221 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 23 b (new)

23b. Calls on the Commission to explore entrusting an existing EU body, such as ENISA, EDPS, or the European Ombudsman, to ensure the harmonised implementation of the European ethical framework for high-risk artificial intelligence, robots and related technologies;

Amendment 222 #

Motion for a resolution
Paragraph 23 b (new)

23b. Calls on the Commission to explore entrusting a new or an existing EU body, such as ENISA, EDPS, or the European Ombudsman, to ensure the harmonised implementation of the European ethical framework for high-risk artificial intelligence, robots and related technologies;

Amendment 223 #

Motion for a resolution
Paragraph 24

24. Believes that such a body, as well as the certification referred to in the following paragraph, would not only benefit the development of Union industry and innovation in that context but also increase the awareness of our citizens regarding the opportunities and risks inherent to these technologies;deleted

Amendment 224 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 24

Amendment 225 #

Motion for a resolution
Paragraph 24

Amendment 226 #

Motion for a resolution
Paragraph 24

Amendment 227 #

Motion for a resolution
Paragraph 24

Amendment 228 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 24

24. Believes that such a body~~, as well as the certification referred to in the following paragraph,~~ would not only benefit the development of Union industry and innovation in that context but also increase the awareness of ~~our~~ citizens regarding the opportunities and risks inherent to ~~these~~ technolog~~ies~~y;

Amendment 229 #

Motion for a resolution
Paragraph 24

24. Believes that such a body~~, as well as the certification referred to in the following paragraph,~~ would not only benefit the development of Union industry and innovation in that context but also increase the awareness of our citizens regarding the opportunities and risks inherent to these technologies;

Amendment 23 #

Motion for a resolution
Recital D

~~D. whereas Parliament has carried out substantial research and adopted several positions on the legal and ethical questions relating to these technologies;~~deleted

Amendment 230 #

Motion for a resolution
Paragraph 24

24. Believes that such a body, as well as the European certification referred to in the following paragraph, would not only benefit the development of Union industry and innovation in that context but also increase the awareness of our citizens regarding the opportunities and risks inherent to these technologies;

Amendment 231 #

Motion for a resolution
Paragraph 24 a (new)

24a. Suggests to create a centre of expertise, bringing together academia, research, industry, and individual experts at Union level, either as an integral part of or associated with such Agency, to foster exchange of knowledge and technical expertise, and to facilitate collaboration throughout the EU and beyond;

Amendment 232 #

Motion for a resolution
Subheading 12

~~European certification of ethical compliance~~deleted

Amendment 233 #

Motion for a resolution
Subheading 12

~~European certification of ethical compliance~~deleted

Amendment 234 #

Motion for a resolution
Paragraph 25

25. Suggests that the European Agency for Artificial Intelligence develops common criteria and an application process relating to the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority;deleted

Amendment 235 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 25

Amendment 236 #

Motion for a resolution
Paragraph 25

25. ~~Suggests that the European Agency for Artificial Intelligence develops common criteria and an application process relating to the granting of~~Calls on the Commission to explore the possibility to develop a European certificateion of ethical compliance ~~following a~~, to be granted at the request ~~by any~~of developer~~, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority~~s of artificial intelligence, robotics and related technologies, which comply with the European ethical framework;

Amendment 237 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 25

25. Suggests that the ~~European Agency for Artificial Intelligence~~Commission, together with the national supervisory authorities, develops common criteria and an application process relating to the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority;

Amendment 238 #

Motion for a resolution
Paragraph 25

25. Suggests th~~at the European Agency for Artificial Intelligence develops common criteria~~e creation of a European certificate of compliance and an application process relating to the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory ~~authority~~system;

Amendment 239 #

Motion for a resolution
Paragraph 25

25. Suggests th~~at the European Agency for Artificial Intelligence develops common criteria and an application process relating to~~e development of common criteria and procedures governing the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority;

Amendment 24 #

Motion for a resolution
Recital D a (new)

Da. whereas the Union has a strict legal framework in place to ensure, inter alia, the protection of personal data and privacy and against discrimination, promote gender balance, environment protection and consumers’ rights;

Amendment 240 #

Motion for a resolution
Paragraph 25

25. Suggests that ~~the European Agency for Artificial Intelligence~~action be taken within the European Commission to develops common criteria and an application process relating to the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify at national level the positive assessment of compliance carried out by the respective national supervisory authority;

Amendment 241 #

Motion for a resolution
Paragraph 25

Amendment 242 #

Motion for a resolution
Paragraph 25 a (new)

25a. Believes that such European certificate of ethical compliance would foster ethics by design throughout the supply chain of AI ecosystems;

Amendment 243 #

Motion for a resolution
Paragraph 25 b (new)

25b. Recommends the certification to be a mandatory prerequisite to the eligibility for public procurement procedures on artificial intelligence and automated decision making systems in the Union;

Amendment 244 #

Motion for a resolution
Paragraph 25 c (new)

25c. Proposes that the European Agency for Artificial Intelligence coordinate the determination of breaches of the principles of non-discrimination by the competent supervisory authorities, in cooperation with other competent authorities in the Union, notably the Consumer Protection Cooperation Network, national consumer protection bodies and civil society, and facilitate means for individuals to meaningfully contest and remedy harm caused by such discrimination, and other infringement of fundamental rights by artificial intelligence and automated decision making systems, whether stemming from public or private sector actors;

Amendment 245 #

Motion for a resolution
Paragraph 26

26. Stresses that the Union’s ethical principles for the development, deployment and use of these technologies ~~should~~could make Europe a world leader in the IA sector and should therefore be promoted worldwide by cooperating with international partners and liaising with third countries with different development and deployment models.

Amendment 246 #

Motion for a resolution
Paragraph 26

26. Stresses that the Union’s legal and ethical principles for the development, deployment and use of these technologies should be promoted worldwide by cooperating with international partners and ~~liaising with third countries with different~~continuing the critical and ethics-based dialogue with third countries implementing alternative models of AI regulation, development and deployment ~~models~~.

Amendment 247 #

Motion for a resolution
Paragraph 26

26. Stresses that the Union’s ethical principles for the development, deployment and use of these high-risk technologies should be promoted worldwide by cooperating with international partners and liaising with third countries with different development and deployment models.

Amendment 248 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 27

27. Recalls that the opportunities and risks inherent to these technologies have a global dimension ~~that requires a consistent approach at international level and thus calls on the Commission to work in bilateral and multilateral settings to advocate and ensur~~, as the software and data used for these systems are frequently imported in and exported out of the European Union; calls on the Commission to take the initiative to investigate which bilateral and multilateral treaties and agreements should be adjusted to ensure a consistent approach and promote ethical compliance globally;

Amendment 249 #

Motion for a resolution
Paragraph 27

27. Recalls that the opportunities and risks inherent to these technologies have a global dimension that requires a consistent approach at international level and thus calls on the Commission to work in bilateral and multilateral settings to advocate and ~~ensure~~promote the European model of ethical compliance.

Amendment 25 #

Motion for a resolution
Recital D b (new)

Db. whereas such an extensive body of horizontal and sectoral legislation , including the existing rules on product safety and liability, will continue to apply in relation to artificial intelligence, robotics and related technologies, although certain adjustments to specific legal instruments may be necessary to reflect the digital transformation and address new challenges posed by the use of artificial intelligence;

Amendment 250 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 27

27. Recalls that the opportunities and risks inherent to these technologies have a global dimension that requires ~~a consistent approach at int~~international coopernation~~al level~~ and thus calls on the Commission to work in bilateral and multilateral settings to advocate and ensure ethical compliance.

Amendment 251 #

Motion for a resolution
Paragraph 28

~~28. Points out the added-value of a European Agency as referred to above in this context as well.~~deleted

Amendment 252 #

Motion for a resolution
Paragraph 28

~~28. Points out the added-value of a European Agency as referred to above in this context as well.~~deleted

Amendment 253 #

Motion for a resolution
Paragraph 28

~~28. Points out the added-value of a European Agency as referred to above in this context as well.~~deleted

Amendment 254 #

Motion for a resolution
Paragraph 28

~~28. Points out the added-value of a European Agency as referred to above in this context as well.~~deleted

Amendment 255 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 28

~~28. Points out the added-value of a European Agency as referred to above in this context as well.~~deleted

Amendment 256 #

Motion for a resolution
Paragraph 28

28. Points out the added-value of a European ~~Agency~~regulatory framework for high- risk artificial intelligence, robotics and related technologies as referred to above in this context as well.

Amendment 257 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 29

Amendment 258 #

Motion for a resolution
Paragraph 29

29. Concludes, following the above reflections on aspects related to the ethical dimension of high-risk artificial intelligence, robotics and related technologies, that the ethical dimension should be framed as a series of principles resulting in a legal framework at Union level ~~supervised by national competent authorities, coordinated and enhanced by a European Agency for Artificial Intelligence and duly respected and certified within the internal market~~;

Amendment 259 #

Motion for a resolution
Paragraph 29

Amendment 26 #

Motion for a resolution
Recital E

E. whereas ~~such questions should be addressed through a comprehensive~~in addition to adjustments to existing legislation a common and future-proof legal framework reflecting the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights ~~that would~~is needed to address the ethical questions related to development, deployment and use of artificial intelligence, robotics and related technologies in order to bring legal certainty to businesses and citizens alike;

Amendment 260 #

Motion for a resolution
Paragraph 29

29. Concludes, following the above reflections on aspects related to the legal and ethical dimensions of artificial intelligence, robotics and related technologies, that the ethical dimension should be ~~framed as a series of principles resulting~~enshrined in a legal framework at Union level, sup~~ervis~~ported by ~~national~~ competent supervisory authorities, coordinated and enhanced by athe aforementioned European Agency for Artificial Intelligence, regularly evaluated by the aforementioned centre of expertise and duly respected and certified within the internal market;

Amendment 261 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 29

29. Concludes, following the above reflections on aspects related to the high- risk ethical dimension of artificial intelligence, robotics and related technologies, that the ethical dimension should be framed as a series of principles resulting in a legal framework at Union level supervised by national competent authorities, ~~coordinated and enhanced by a European Agency for Artificial Intelligence~~ and duly respected and certified within the internal market;

Amendment 262 #

Motion for a resolution
Paragraph 29

Amendment 263 #

Motion for a resolution
Paragraph 29

29. Concludes, following the above reflections on aspects related to the ethical dimension of artificial intelligence, robotics and related technologies, that the ethical dimension should be framed as a series of principles resulting in a legal framework at Union level supervised by national competent authorities, coordinated and enhanced by a European Agency for Artificial Intelligence and duly respected and certified within the internal marketn effective, forward looking and comprehensive legal framework at Union level, reflecting the standards and values enshrined in the treaties and in the Charter of Fundamental Rights and affording protection, trust and legal safety to members of the public and economic operators;

Amendment 264 #

Motion for a resolution
Paragraph 29

Amendment 265 #

Motion for a resolution
Paragraph 30

30. Following the procedure of Article 225 of the Treaty on the Functioning of the European Union, requests the Commission to submit a proposal for a Regulation on ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies on the basis of Article 114 of the Treaty on the Functioning of the European Union and following the detailed recommendations set out in the annex hereto, and the proposal should not undermine sector-specific legislation, but only cover the legal loopholes;

Amendment 266 #

Motion for a resolution
Paragraph 30

Amendment 267 #

Motion for a resolution
Paragraph 30

30. Following the procedure of Article 225 of the Treaty on the Functioning of the European Union, requests the Commission to submit a proposal for a Regulation on ethical principles for the development, deployment and use of high-risk artificial intelligence, robotics and related technologies on the basis of Article 114 of the Treaty on the Functioning of the European Union and following the detailed recommendations set out in the annex hereto;

Amendment 268 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 31

31. Recommends that the European Commission review existing Union law applicable to artificial intelligence, robotics and related technologies in order to address the rapidity of their development in line with the recommendations set out in the annex hereto, avoiding both over- regulation and administrative burdens, especially for SMEs;

Amendment 269 #

Motion for a resolution
Paragraph 31

31. Recommends that the European Commission, after consulting with all the relevant stakeholders, review existing Union law applicable to high-risk artificial intelligence, robotics and related technologies in order to review it when necessary and address the rapidity of their development in line with the recommendations set out in the annex hereto;

Amendment 27 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital E

E. whereas ~~such question~~legal and ethical questions relating to these technologies should be addressed through a comprehensive and future-proof ~~legal~~ framework of Union law, reflecting the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights t~~hat would bring~~o increase legal certainty tofor businesses and citizens alike and to secure civil rights, labour rights and environmental standards;

Amendment 270 #

Motion for a resolution
Paragraph 31 a (new)

31a. Believes that a periodical assessment of the European regulatory framework related to artificial intelligence, robotics and related technologies will be essential to ensure that the applicable legislation is up to date with the rapidly growing technological progress;

Amendment 271 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Paragraph 32

~~32. Considers that the requested proposal would have financial implications if a new European Agency for Artificial Intelligence is set up;~~deleted

Amendment 272 #

Motion for a resolution
Paragraph 32

~~32. Considers that the requested proposal would have financial implications if a new European Agency for Artificial Intelligence is set up;~~deleted

Amendment 273 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 32

~~32. Considers that the requested proposal would have financial implications if a new European Agency for Artificial Intelligence is set up;~~deleted

Amendment 274 #

Motion for a resolution
Paragraph 32

32. Considers that the requested proposal would have financial implications if a ~~new European Agency for Artificial Intelligence is set up~~n existing European body is entrusted with the above-mentioned functions in order to ensure the necessary technical means and human resources to fulfil its newly attributed tasks;

Amendment 275 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 32

32. Considers that the requested proposal ~~would~~does not have financial implications ~~if a new European Agency for Artificial Intelligence is set up~~;

Amendment 276 #

Motion for a resolution
Annex I – part A – point I – indent 1

- to build trust in artificial intelligence, robotics and related technologies by ensuring that when these technologies ~~will be~~entail a high-risk for the protection of safety and fundamental rights and freedoms their developedment, deployedment and used will be done in an ethical manner;

Amendment 277 #

Motion for a resolution
Annex I – part А – point I – indent 1

- to build trust in artificial intelligence, robotics and related technologies by ensuring that these technologies will be developed, deployed and used in ~~an ethical manner~~full compliance with the EU’s ethical principles in the field of artificial intelligence;

Amendment 278 #

Motion for a resolution
Annex I – part A – point I – indent 1

- to build trust in ~~artificial intelligence, robotics and related~~intelligent and autonomous technologies by ensuring that these technologies will be developed, deployed and used in an ethical manner at all levels of involved stakeholders and of society;

Amendment 279 #

Motion for a resolution
Annex I – part A – point I – indent 1

- to build trust in high-risk artificial intelligence, robotics and related technologies by ensuring that these technologies will be developed, deployed and used in an ethical manner;

Amendment 28 #

Motion for a resolution
Recital E

E. whereas such questions should be addressed through a comprehensive and future-proof legal framework reflecting the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights that would bring legal certainty to businesses and citizens alike and at the same time refraining from over-regulation by only closing existing legal loopholes;

Amendment 280 #

Motion for a resolution
Annex I – part A – point I – indent 1 a (new)

- to ensure the balance between public protection on the one hand and business incentives to invest in innovation, especially AI systems, on the other;

Amendment 281 #

Motion for a resolution
Annex I – part A – point I – indent 2

- to support the development of artificial intelligence, robotics and related technologies in the Union, including by helping businesses and start-ups to assess and address with certainty current and future regulatory requirements and risks during the ~~development proces~~innovation and business development process, and during the subsequent phase of use by professionals and private individuals;

Amendment 282 #

Motion for a resolution
Annex I – part A – point I – indent 2

- to support the development of artificial intelligence, robotics and related technologies in the Union, including by helping businesses ~~and~~, start-ups and SMEs to assess and address regulatory requirements and risks during the development process, especially by minimising burdens and red tape;

Amendment 283 #

Motion for a resolution
Annex I – part A – point I – indent 2

- to support the development of artificial intelligence, robotics and related technologies in the Union, including by helping businesses, small and medium- sized enterprises and start-ups to assess and address regulatory requirements and risks during the development process;

Amendment 284 #

Motion for a resolution
Annex I – part A – point I – indent 2

- to support the development of high- risk artificial intelligence, robotics and related technologies in the Union, including by helping businesses and start- ups to assess and address regulatory requirements and risks during the development process;

Amendment 285 #

Motion for a resolution
Annex I – part A – point I – indent 3

- to support deployment of artificial intelligence, robotics and related technologies in the Union by providing the appropriate regulatory framework, in the aim of encouraging regulatory certainty and innovation while guaranteeing fundamental rights and consumer protection;

Amendment 286 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point I – indent 3

- to support deployment of artificial intelligence, robotics and related technologies in the Union by providing the appropriate regulatory framework which should apply as a complement to existing sector-specific legislation;

Amendment 287 #

Motion for a resolution
Annex I – part A – point I – indent 3

- to support deployment of artificial intelligence, robotics and related technologies in the Union by providing the appropriate and proportionate regulatory framework;

Amendment 288 #

Motion for a resolution
Annex I – part A – point I – indent 3

- to support deployment of high-risk artificial intelligence, robotics and related technologies in the Union by providing the appropriate regulatory framework;

Amendment 289 #

Motion for a resolution
Annex I – part A – point I – indent 4

~~- to support use of artificial intelligence, robotics and related technologies in the Union by ensuring that they are developed, deployed and used in an ethical manner;~~deleted

Amendment 29 #

Motion for a resolution
Recital E

E. whereas such questions should be addressed through an effective, comprehensive and future-proof legal framework reflecting the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights that would bring protection, confidence and legal certainty to businesses and citizens alike;

Amendment 290 #

Motion for a resolution
Annex I – part A – point I – indent 4

- to support use of high-risk artificial intelligence, robotics and related technologies in the Union by ensuring that they are developed, deployed and used in an ethical manner;

Amendment 291 #

Motion for a resolution
Annex I – part A – point I – indent 5

- to require transparency and better information flows among citizens and within organisations developing, deploying or using artificial intelligence, robotics and related technologies as a means of ensuring that these technologies are compliant with Union law, human rights and values, and with the ethical principles of the proposed Regulation.

Amendment 292 #

Motion for a resolution
Annex I – part A – point I – indent 5

- to require better information flows among citizens and within organisations developing, deploying or using high-risk artificial intelligence, robotics and related technologies as a means of ensuring that these technologies are compliant with the ethical principles of the proposed Regulation.

Amendment 293 #

Motion for a resolution
Annex I – part A – point II – indent 1

- a “Regulation on ethical principles for the development, deployment and use of high-risk artificial intelligence, robotics and related technologies”;

Amendment 294 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point II – indent 2

~~- a European Agency for Artificial Intelligence and a European certification of ethical compliance;~~deleted

Amendment 295 #

Motion for a resolution
Annex I – part A – point II – indent 2

~~- a European Agency for Artificial Intelligence and a European certification of ethical compliance;~~deleted

Amendment 296 #

Motion for a resolution
Annex I – part A – point II – indent 2

~~- a European Agency for Artificial Intelligence and a European certification of ethical compliance;~~deleted

Amendment 297 #

Motion for a resolution
Annex I – part A – point II – indent 2

- ~~a European Agency for Artificial Intelligence and~~ a European certification of ethical compliance;

Amendment 298 #

Motion for a resolution
Annex I – part A – point II – indent 4

- the work carried out by the “Scompetent supervisory Aauthority”ies in ~~each~~the Member States to ensure that ethical principles are applied to artificial intelligence, robotics and related technologies;

Amendment 299 #

Motion for a resolution
Annex I – part A – point II – indent 4

- the work carried out by the “Supervisory Authority” in each Member State to ensure that ethical principles are applied to high-risk artificial intelligence, robotics and related technologies;

Amendment 3 #

Motion for a resolution
Citation 2 b (new)

- having regard to Directive (EU) 2019/770 of the European Parliament and of the Council of 20 May 2019 on certain aspects concerning contracts for the supply of digital content and digital services,

Amendment 30 #

Motion for a resolution
Recital E

E. whereas ~~such questions should be addressed through a comprehensive and future-proof legal framework~~there are new issues arising related to the existing legal framework; whereas the legal framework should be future-proof and should reflect~~ing~~ the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights that would bring legal certainty to businesses and citizens alike;

Amendment 300 #

Motion for a resolution
Annex I – part A – point II – indent 5

- the involvement and consultation of, as well as provision of support to, relevant research and development, and concerned stakeholders, including start- ups, businesses, social partners, and other representatives of the civic society.

Amendment 301 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point II – indent 5

- the involvement and consultation of, as well as provision of support to, stakeholders, including start-ups, small and medium-sized enterprises, businesses, social partners, and other representatives of the civic society.

Amendment 302 #

Motion for a resolution
Annex I – part A – point II – indent 5 a (new)

- annex establishing a list of high- risk technologies which fall under the scope of this Regulation;

Amendment 303 #

Motion for a resolution
Annex I – part A – point III – introductory part

III. The “Regulation on ethical principles for the development, deployment and use of high-risk artificial intelligence, robotics and related technologies” builds on the following principles:

Amendment 304 #

Motion for a resolution
Annex I – part А – point III – indent 1

- hHuman-centric, human-made and human-~~made~~controlled artificial intelligence, robotics and related technologies;

Amendment 305 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point III – indent 1

- human-centric ~~and human-made~~ artificial intelligence, robotics and related technologies;

Amendment 306 #

Motion for a resolution
Annex I – part A – point III – indent 2

- ~~risk~~compliance assessment of high- risk artificial intelligence, robotics and related technologies;

Amendment 307 #

Motion for a resolution
Annex I – part A – point III – indent 3

- safety features, transparency ~~and~~, accountability, and accessibility;

Amendment 308 #

Motion for a resolution
Annex I – part A – point III – indent 4

- safeguards and remedies against bias and discrimination;

Amendment 309 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Annex I – part A – point III – indent 4

- safeguards against unfair bias and discrimination;

Amendment 31 #

Motion for a resolution
Recital E

E. whereas such questions should be addressed through a comprehensive and future-proof legal framework reflecting the Union’s principles and values as enshrined in the Treaties and the Charter of Fundamental Rights that would bring legal certainty to ~~businesses and citizen~~all citizens and businesses alike;

Amendment 310 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point III – indent 4 a (new)

- right to redress;

Amendment 311 #

Motion for a resolution
Annex I – part A – point III – indent 5

- social responsibility and gender ~~balance~~equality in artificial intelligence, robotics and related technologies;

Amendment 312 #

Sergey LAGODINSKY, Alexandra GEESE, Patrick BREYER

Motion for a resolution
Annex I – part A – point III – indent 6

- environmentally ~~friendly and~~ sustainable artificial intelligence, robotics and related technologies;

Amendment 313 #

Motion for a resolution
Annex I – part A – point III – indent 7

- respect for fundamental rights, including privacy and limitations to the use of biometric recognition;

Amendment 314 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point III – indent 7 a (new)

- safeguards related to the use of high-risk artificial intelligence, robotics and related technologies within the framework of public power decisions;

Amendment 315 #

Motion for a resolution
Annex I – part A – point IV – introductory part

IV. The key elements of the Commission’s task as regards compliance with ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies are: - to issue guidance as regards the application of the proposed regulation; - to develop a European certificate of compliance with ethical principles; - to support regular exchanges with concerned stakeholders and the civil society.

Amendment 316 #

Motion for a resolution
Annex I – part A – point IV – introductory part

IV. The key elements of the Commission’s task as regards compliance with legal and ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies are:

Amendment 317 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point IV – indent 1 a (new)

- to issue guidance as regards the application of the proposed Regulation in order to ensure its consistent application, namely regarding the application of the criteria for artificial intelligence, robotics and related technologies to be considered high-risk;

Amendment 318 #

Motion for a resolution
Annex I – part A – point IV – indent 1 a (new)

- regularly assessing and if necessary, reviewing the annex of the Regulation by means of a delegated act

Amendment 319 #

Motion for a resolution
Annex I – part A – point IV – indent 1 b (new)

- to liaise with the “Supervisory Authority” in each Member State;

Amendment 32 #

Motion for a resolution
Recital E a (new)

Ea. Whereas the diversity of applications driven by artificial intelligence, robotics and related technologies complicates finding a single solution suitable for the entire spectrum of risks; notes in this respect that an approach should be adopted in which experiments, pilots and regulatory sandboxes are used to come up with proportional and evidence-based solutions;

Amendment 320 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point IV – indent 2

- raising awareness, providing information, education, training and engaging in exchanges with designers, developers, deployers ~~and users throughout the Union~~, citizens, users and institutional bodies throughout the Union and internationally.

Amendment 321 #

Motion for a resolution
Annex I – part A – point IV – indent 2 a (new)

- engaging discussions on global ethical norms at international level;

Amendment 322 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point IV – indent 2 b (new)

- establishing binding guidelines on the methodology of the compliance assessment to be followed by the national supervisory authorities;

Amendment 323 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point IV – indent 2 c (new)

- establishing non-binding guidelines directed to the developers, the deployers and the users;

Amendment 324 #

Motion for a resolution
Annex I – part A – point V

V. The European Agency for Artificial Intelligence should be established following a detailed proposal from the Commission, which should include the following main tasks: - to supervise the application of the proposed Regulation; - to issue guidance as regards the application of the proposed Regulation; - to liaise with the “Supervisory Authority” in each Member State and coordinate their mandate and tasks; - to develop a European certificate of compliance with ethical principles; - concerned stakeholders and the civil society.deleted to support regular exchanges with

Amendment 325 #

Motion for a resolution
Annex I – part A – point V

V. The European Agency for Artificial Intelligence should be established following a detailed proposal from the Commission, which should include the following main tasks: - to supervise the application of the proposed Regulation; - to issue guidance as regards the application of the proposed Regulation; - to liaise with the “Supervisory Authority” in each Member State and coordinate their mandate and tasks; - of compliance with ethical principles; - concerned stakeholders and the civil society.deleted to develop a European certificate to support regular exchanges with

Amendment 326 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point V

V. The European Agency for Artificial Intelligence should be established following a detailed proposal from the Commission, which should include the following main tasks: - proposed Regulation; - application of the proposed Regulation; - Authority” in each Member State and coordinate their mandate and tasks; - of complideleted to supervise the application of the to issue guidance as regards the to liaise with the “Supervisory to develop a European ce ~~with ethical principles; - concerned stakeholders and the civil society.~~rtificate to support regular exchanges with

Amendment 327 #

Motion for a resolution
Annex I – part A – point V

Amendment 328 #

Motion for a resolution
Annex I – part A – point V – indent 1

~~- to supervise the application of the proposed Regulation;~~deleted

Amendment 329 #

Motion for a resolution
Annex I – part A – point V – indent 1

- to supervise the application of ~~the proposed Regulation~~relevant Union law;

Amendment 33 #

Motion for a resolution
Recital E b (new)

Eb. Whereas the future regulatory framework needs to take into consideration all the interests at stake; whereas careful examination of the consequences of any new regulatory framework on all actors in an impact assessment should be a prerequisite for further legislative steps; whereas the crucial role of Small- and Medium sized enterprises (SMEs) and start-ups especially in the European economy justifies a strictly proportionate approach to enable them to develop and innovate;

Amendment 330 #

Motion for a resolution
Annex I – part A – point V – indent 2

~~- to issue guidance as regards the application of the proposed Regulation;~~deleted

Amendment 331 #

Motion for a resolution
Annex I – part A – point V – indent 2

- to issue guidance as regards the application of ~~the proposed Regulation~~relevant Union law;

Amendment 332 #

Motion for a resolution
Annex I – part A – point V – indent 3

~~- to liaise with the “Supervisory Authority” in each Member State and coordinate their mandate and tasks;~~deleted

Amendment 333 #

Motion for a resolution
Annex I – part A – point V – indent 3

- to liaise with the “SMember States and concerned competent supervisory Aauthorit~~y” in each Member State and coordinate their mandate and tasks~~ies, coordinate their contributions and collaboration;

Amendment 334 #

Motion for a resolution
Annex I – part A – point V – indent 3 a (new)

- to draft a common framework for the governance of the development, deployment and use of artificial intelligence, robotics and related technologies to be implemented by the national supervisory authorities;

Amendment 335 #

Motion for a resolution
Annex I – part A – point V – indent 4

~~- to develop a European certificate of compliance with ethical principles;~~deleted

Amendment 336 #

Motion for a resolution
Annex I – part A – point V – indent 4

- to develop a European certificate of compliance with legal and ethical principles with specific requirements for artificial intelligence, robotics and related technologies targeting children, young adults or other vulnerable persons;

Amendment 337 #

Motion for a resolution
Annex I – part A – point V – indent 5

~~- to support regular exchanges with concerned stakeholders and the civil society.~~deleted

Amendment 338 #

Motion for a resolution
Annex I – part A – point V – indent 5

- to support ~~regular exchanges with concerned stakeholders and the civil society~~cross-sector and cross- border co-operation through regular exchanges, in the form of a structured dialogue, with concerned stakeholders and the civil society, in the EU and in the world, notably with social partners, researchers and competent authorities in the fields of neuroscience, cognitive science, health, sociology, economy and law.

Amendment 339 #

Motion for a resolution
Annex I – part A – point V – indent 5 a (new)

- to support regular exchanges with concerned stakeholders as regards the development of technical standards at international level.

Amendment 34 #

Motion for a resolution
Recital F

F. whereas for the scope of that framework to be adequate it should cover a wide range of technologies and their components, including algorithms, software and data used or produced by them, but it should be borne in mind that, due to these technologies, this regulation should only be used in addition to the existing sector-specific legislation;

Amendment 340 #

Motion for a resolution
Annex I – part A – point V – indent 5 b (new)

- to organise or support the activities of the centre of experts, assessing technological standards and evaluation this regulation and work of the Agency based on expertise of their members in the relative realm of technology and humanities.

Amendment 341 #

Motion for a resolution
Annex I – part A – point VI – introductory part

VI. The key tasks of ~~the “S~~competent supervisory Aauthority”ies in ~~each~~the Member States should be:

Amendment 342 #

Motion for a resolution
Annex I – part A – point VI – indent 1

- to assess whether artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed and used in the Union are high-risk technologies in accordance with the criteria defined in the proposed Regulation;

Amendment 343 #

Motion for a resolution
Annex I – part A – point VI – indent 1

- to assess wthe~~ther~~ compliance of high- risk artificial intelligence, robotics and related technologies~~, including software, algorithms and data u~~ with the ethical principles sedt o~~r produced by such technologies, developed, deployed and used in the Union are high-risk technologies~~ut in the proposed Regulation;

Amendment 344 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point VI – indent 1

Amendment 345 #

Motion for a resolution
Annex I – part A – point VI – indent 2

~~- to monitor their compliance with the ethical principles set out in the proposed Regulation;~~deleted

Amendment 346 #

Motion for a resolution
Annex I – part A – point VI – indent 2

- to monitor their compliance with ~~the~~law and ethical principles set out in the proposed Regulation;

Amendment 347 #

Motion for a resolution
Annex I – part A – point VI – indent 2 a (new)

- to issue a certificate of compliance with ethical principles, in line with common criteria and an application process developed in cooperation with other Supervisory Authorities, the European Commission and other relevant institutions, bodies, offices and agencies of the Union;

Amendment 348 #

Motion for a resolution
Annex I – part A – point VI – indent 3

- to contribute to the consistent application of the proposed Regulation in cooperation with other Supervisory Authorities, the European Commission and other relevant institutions, bodies, offices and agencies of the Union, namely regarding the application of the criteria for artificial intelligence, robotics and related technologies to be considered high-risk by elaborating, in the context of such cooperation, a common and exhaustive list of high-risk artificial intelligence, robotics and related technologies in line with the criteria set out in this Regulation; and

Amendment 349 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point VI – indent 3

- to contribute to the consistent application of the legal framework and the proposed Regulation in cooperation with other Scompetent supervisory Aauthorities, the European Commission and other relevant institutions, bodies, offices and agencies of the Union; and

Amendment 35 #

Motion for a resolution
Recital F

F. whereas ~~for~~ the scope of that framework toshould be adequate, proportionate and thoroughly assessed; whereas while it should cover a wide range of technologies and their components, including algorithms, software and data used or produced by them, a targeted approach based on the concept of high risk is necessary to avoid hampering future innovations;

Amendment 350 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part A – point VI – indent 4

- to be responsible for establishing standards for the governance of artificial intelligence, robotics and related technologies, including by liaising with the maximum possible number of stakeholders and civil society representatives.deleted

Amendment 351 #

Motion for a resolution
Annex I – part A – point VI – indent 4

- to be responsible for establishing standards for the governance of artificial intelligence, robotics and related technologies, including by lisustaisning ~~with the maximum possible number of~~a structured dialogue with all relevant stakeholders and civil society representatives.;

Amendment 352 #

Motion for a resolution
Annex I – part A – point VI – indent 4

- to be responsible for ~~establish~~implementing standards for the governance of artificial intelligence, robotics and related technologies, including by liaising with the maximum possible number of stakeholders and civil society representatives.

Amendment 353 #

Motion for a resolution
Annex I – part A – point VI – indent 4 a (new)

- to cooperate with the authorities responsible for implementing sectorial legislation in order to identify technologies which are high-risk from an ethical perspective and in order to supervise the implementation of required and appropriate measures where such technologies are identified;

Amendment 354 #

Motion for a resolution
Annex I – part A – point VI – indent 4 a (new)

- to raise awareness, provide information on artificial intelligence, robotics and related technologies to the public, and support the training of relevant professions, including in the judiciary;

Amendment 355 #

Motion for a resolution
Annex I – part A – point VI – indent 4 a (new)

- to support regular exchanges with concerned stakeholders and civil society.

Amendment 356 #

Motion for a resolution
Annex I – part A – point VI – indent 4 b (new)

- to empower people with the skills for AI and support workers for a fair transition;

Amendment 357 #

Motion for a resolution
Annex I – part A – point VI – indent 4 c (new)

- to serve as a first point of contact in cases of suspected discriminating treatment or of violation of other rights as a result of use of artificial intelligence and automated decision making systems, conduct ethical evaluation of such cases in cooperation with other competent authorities in the Union, notably the Consumer Protection Cooperation Network, national consumer protection bodies and civil society; to facilitate means to individuals to meaningfully contest and remedy harm caused by such discrimination or violations;

Amendment 358 #

Motion for a resolution
Annex I – part A – point VI – indent 4 d (new)

- to facilitate the sharing of best practice and techniques for de-biasing datasets, and combatting discrimination in artificial intelligence and automated decision making systems, between competent supervisory authorities, research and development, civil society and individual experts.

Amendment 359 #

Motion for a resolution
Annex I – part A – point VII

VII. The key role of stakeholders should be to engage with the Commission~~, the European Agency for Artificial Intelligence~~ and the “Supervisory Authority” in each Member State.

Amendment 36 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital F

F. whereas for the scope of ~~that~~such regulatory framework to be adequate ~~it should cover a wide range of technologies and their components, including algorithms, software and data used or produced by them~~, proportionate and avoid the creation of unnecessary burdens, especially for SMEs, it should cover only those technologies in relation to which, considering their intended use and the sectors where they are employed, significant risk can be expected to occur;

Amendment 360 #

Motion for a resolution
Annex I – part A – point VII

VII. The key role of stakeholders should be to engage with the Commission~~, the European Agency for Artificial Intelligence~~ and the “Supervisory Authority” in each Member State.

Amendment 361 #

Motion for a resolution
Annex I – part A – point VII

VII. The key role of stakeholders should be to engage with the Commission~~, the European Agency for Artificial Intelligence~~ and the “Supervisory Authority” in each Member State.

Amendment 362 #

Motion for a resolution
Annex I – part A – point VII

VII. The key role of stakeholders should be to engage with the Commission, the European Agency for Artificial Intelligence and the “Scompetent supervisory Aauthority”ies in ~~each~~ Member States.

Amendment 363 #

Motion for a resolution
Annex I – part A – point VII

VII. The key role of stakeholders should be to engage with the ~~Commission, the~~relevant, new or existing, European Agency for ~~Artificial Intelligence~~European body and the “Supervisory Authority” in each Member State.

Amendment 364 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 1

(1) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, are based on a desire to serve society. They can entail opportunities and risks, which should be addressed and regulated by an efficient, forward- looking and comprehensive legal framework of ethical principles to be complied with from the moment of the development and deployment of such technologies to their use.

Amendment 365 #

Motion for a resolution
Annex I – part B – recital 1

(1) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, arshould be based on a desire to serve society. ~~They~~Some of these technologies can entail opportunities and risks, which should be addressed and regulated by a comprehensive legal framework of ethical principles to be complied with from the moment of the development and deployment of such technologies to their use.

Amendment 366 #

Motion for a resolution
Annex I – part B – recital 1

(1) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, are based on a desire to serve society. They can entail opportunities and risks, which should be addressed and regulated by a comprehensive legal framework of ethical principles to be complied with by high-risk technologies from the moment of the development and deployment of such technologies to their use.

Amendment 367 #

Motion for a resolution
Annex I – part B – recital 1

(1) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, ~~are based on a desire to~~should serve society. They can entail opportunities and risks, which should be addressed and regulated by a comprehensive legal framework of Union law, reflecting ethical principles to be complied with from the moment of the development and deployment of such technologies to their use.

Amendment 368 #

Motion for a resolution
Annex I – part B – recital 2

(2) The level of compliance with the ethical principles regarding the development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies in the Union should be equivalent in all Member States, in order to efficiently seize the opportunities, avoid regulatory fragmentation and consistently address the risks of such technologies. It should be ensured that the application of the rules set out in this Regulation throughout the Union is homogenous.

Amendment 369 #

Motion for a resolution
Annex I – part B – recital 2

(2) The level of compliance with the ethical principles regarding the development, deployment and use of high- risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies in the Union should be equivalent in all Member States, in order to efficiently seize the opportunities and consistently address the risks of such technologies. It should be ensured that the application of the rules set out in this Regulation throughout the Union is homogenous.

Amendment 37 #

Motion for a resolution
Recital F

F. whereas for the scope of thate legal framework to be adequate it should cover a wide range of technologies and their components, including algorithms, software and data used or produced by them, but also the global technical infrastructures allowing these technologies to function;

Amendment 370 #

Motion for a resolution
Annex I – part B – recital 3

(3) In this context, the current diversity of the rules and practices to be followed across the Union poses a significant risk of fragmentation of the single market and to the protection of the well-being and prosperity of individuals and society alike, as well as to the coherent exploration of the full potential that artificial intelligence, robotics and related technologies have in promoting and preserving that well-being and prosperity. Differences in the degree of consideration of the ethical dimension inherent to these technologies can prevent them from being freely developed, deployed or used within the Union and such differences can constitute an obstacle to the pursuit of economic activities at Union level, distort competition and impede authorities in the fulfilment of their obligations under Union law. In addition, the absence of a common framework of ethical principles for the development, deployment and use of high-risk artificial intelligence, robotics and related technologies results in legal uncertainty for all those involved, namely developers, deployers and users.

Amendment 371 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 3

(3) In this context, the current diversity of the rules and practices to be followed across the Union poses a significant risk to the protection of the well-being and prosperity of individuals and society alike, as well as to the coherent exploration of the full potential that artificial intelligence, robotics and related technologies have in promoting and preserving that well-being and prosperity. Differences in the degree of consideration of the ethical dimension inherent to these technologies can prevent them from being freely developed, deployed or used within the Union and such differences can constitute an obstacle to the pursuit of economic activities at Union level, distort competition and impede authorities in the fulfilment of their obligations under Union law. In addition, the absence of a common legal framework of, reflecting the Union’s ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies results in legal uncertainty for all those involved, namely developers, deployers and users.

Amendment 372 #

Motion for a resolution
Annex I – part B – recital 3

(3) In this context, the current diversity of the rules and practices to be followed across the Union poses a significant risk ~~to the protection of the well-being and prosperity of individuals and society alike~~of fragmentation of the Single Market , as well as to the coherent exploration of the full potential that artificial intelligence, robotics and related technologies have in promoting ~~and preserving that well-being and prosperity~~innovation. Differences in the degree of consideration of the ethical dimension inherent to these technologies can prevent them from being freely developed, deployed or used within the Union and such differences can constitute an obstacle to ~~the pursuit of~~a level playing field and to the pursuit of technological progress and economic activities at Union level, distort competition and impede authorities in the fulfilment of their obligations under Union law. In addition, the absence of a common framework of ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies results in legal uncertainty for all those involved, namely developers, deployers and users.

Amendment 373 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 4

(4) Nevertheless, this Regulation should provide a margin of manoeuvre for Member States, including with regard to how the mandate of their respective national supervisory authority is to be carried out in view of the objectives it is to pursue as laid down herein.deleted

Amendment 374 #

Motion for a resolution
Annex I – part B – recital 4

(4) Nevertheless, this Regulation should provide a margin of manoeuvre for Member States, including with regard to how the mandate of their ~~respective national~~competent supervisory authorityies is to be carried out in view of the objectives it is to pursue as laid down herein.

Amendment 375 #

Motion for a resolution
Annex I – part B – recital 4 a (new)

(4a) In addition, this Regulation should only supplement existing sector-specific legislation and not undermine it. At the same time, red tape must be reduced as far as possible and the burdens minimised, especially for SMEs.

Amendment 376 #

Motion for a resolution
Annex I – part B – recital 5

(5) The ~~geographical~~ scope of application of such a framework should cover all the components of artificial intelligence, robotics and related technologies throughout their developedment, deploy~~ed or~~ment and used in the Union, including in cases where part of the technologies might be located outside the Union or not have a s~~pecific~~ingle location, such as in the case of cloud computing services.

Amendment 377 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 6

(6) A common understanding in the Union of notions such as artificial intelligence, robotics, related technologies, algorithms and biometric recognition is required in order to allow for a harmonized regulatory approach. ~~However, the specific legal definitions need to be developed in the context of this Regulation without prejudice to other definitions used in other legal acts and international jurisdictions.~~

Amendment 378 #

Motion for a resolution
Annex I – part B – recital 6

(6) A common understanding in the Union of notions such as artificial intelligence, robotics, related technologies, algorithms and biometric recognition is required in order to allow for a harmonized regulatory approach. However, the specific legal definitions need to be developed in the context of this Regulation without prejudice to other definitions used in other legal acts and international jurisdictions. They should be technologically neutral and also subject to review whenever necessary.

Amendment 379 #

Motion for a resolution
Annex I – part B – recital 6

(6) A common understanding in the Union of notions such as artificial intelligence, robotics, related technologies, algorithms and biometric recognition is required in order to allow for a harmonized regulatory approach and thus legal security for citizens and companies. . However, the specific legal definitions need to be developed in the context of this Regulation without prejudice to other definitions used in other legal acts and international jurisdictions.

Amendment 38 #

Motion for a resolution
Recital G

G. whereas thate Union’s legal framework should encompass all ~~situations requiring due consideration of the Union’s principles and valu~~relevant stages, namely from the development, and deployment ~~and~~to the use of the relevant technologies and their components;

Amendment 380 #

Motion for a resolution
Annex I – part B – recital 6

(6) A common understanding in the Union of notions such as artificial intelligence, robotics, related technologies, and algorithms ~~and biometric recognition~~ is required in order to allow for a harmonized regulatory approach. However, the specific legal definitions need to be developed in the context of this Regulation without prejudice to other definitions used in other legal acts and international jurisdictions.

Amendment 381 #

Motion for a resolution
Annex I – part B – recital 6 a (new)

(6a) There are technologies related to artificial intelligence and robotics that enable software to control physical or virtual processes, at a varying degree of autonomy1a. __________________ 1aFor automated driving of vehicles, six levels of driving automation have been proposed by SAE International standard J3016, last updated in 2018 to J3016_201806. https://www.sae.org/standards/content/j30 16_201806/

Amendment 382 #

Motion for a resolution
Annex I – part B – recital 6 b (new)

(6b) Bias can originate both from decisions informed or made by an automated system as well as from data sets on which such decision making is based or trained. Such bias usually signifies a personal or social prejudice or perception of a person or a group based on their real or ascribed traits.

Amendment 383 #

Brando BENIFEI

Motion for a resolution
Annex I – part B – recital 7

(7) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should be ~~such as to~~meant to complement human capabilities, not to substitute them, it should be aimed at ensureing that the best interests of citizens are considered, and it should respect fundamental rights as set out in the Charter of Fundamental Rights of the European Union (‘the Charter’), settled case-law of the Court of Justice of the European Union, and other European and international instruments which apply in the Union.

Amendment 384 #

Motion for a resolution
Annex I – part B – recital 7

(7) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should be such as to ensure that their execution does not run against the best interests of affected citizens a~~re considered,~~nd the public and should respect Union law, based on fundamental rights as set out in the Charter of Fundamental Rights of the European Union (‘the Charter’), settled case-law of the Court of Justice of the European Union, and other European and international instruments which apply in the Union.

Amendment 385 #

Motion for a resolution
Annex I – part B – recital 7

(7) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should be such as to ensure that the best interests of citizens are considered, ~~and should~~by respecting fundamental rights as set out in the Charter of Fundamental Rights of the European Union (‘the Charter’), settled case-law of the Court of Justice of the European Union, and other European and international instruments which apply in the Union.

Amendment 386 #

Motion for a resolution
Annex I – part B – recital 8

(8) Artificial intelligence, robotics and related technologies have been provided with the ability to learn from data and experience, as well as to take founded decisions. Such capacities need to remain subject to meaningful human review, judgment, intervention and control. The technical and operational complexity of such technologies should never prevent their deployer or user from being able to, at the very least, ~~alter or halt~~enable a fail-safe shutdown, alter, halt or redress them in cases where the compliance with the principles set out in this Regulation is at risk.

Amendment 387 #

Motion for a resolution
Annex I – part B – recital 8

(8) AHigh-risk artificial intelligence, robotics and related technologies have been provided with the ability to learn from data and experience, as well as to take founded decisions. Such capacities need to remain subject to meaningful human review, judgment, intervention and control. The technical and operational complexity of such technologies should never prevent their deployer or user from being able to, at the very least, ~~alter or~~ halt them in cases where the compliance with the principles set out in this Regulation is at risk.

Amendment 388 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 8

(8) ADecisions made or informed by artificial intelligence, robotics and related technologies ~~have been provided with the ability to learn from data and experience, as well as to take founded decisions. Such capacities need to~~should remain subject to meaningful human review, judgment, intervention and control. The technical and operational complexity of such technologies should never prevent their deployer or user from being able to, at the very least, alter or halt them in cases where the ~~compliance with the principles set out in~~execution of such technology amounts to essential risk for goods and rights of the user or run counter to the principles of the Union law and this Regulation ~~is at risk~~.

Amendment 389 #

Motion for a resolution
Annex I – part B – recital 8 a (new)

(8a) This Regulation should be strictly proportionate to its objective so as not to hamper innovation in the Union. In this respect, it should be based on a targeted risk-based approach focusing on specific sectors where major interests are at stake.

Amendment 39 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Recital G

G. whereas that framework should ~~encompass all situations requiring due consideration of the Union’s principles and values, namely~~make sure that the development, the deployment and the use of ~~the~~ relevant technologies and their components are fully compliant with the Union’s principles and values;

Amendment 390 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 8 b (new)

(8b) The scope of the Regulation should be limited to high-risk artificial intelligence, robotics and related technologies used in sectors, where given the characteristics of the activities typically undertaken, significant risks can be expected to occur, and where their use is likely to create significant risk. A significant risk should be understood as directly endangering the protection of safety or fundamental rights and freedoms.

Amendment 391 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 8 c (new)

(8c) The sectors covered should be notably, but not exclusively, healthcare, transport, energy and finance. The high- risk technologies should be exclusively listed in the annex of this Regulation, which should be revised on a regular basis, keeping up with technological development.

Amendment 392 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 9

(9) ~~Any artificial intelligence, robotics and related~~High-risk technologies, including the software, algorithms and data used or produced by such technologies, ~~which entails a high risk of breaching~~should respect the principles of safety, transparency, accountability, non-bias or non- discrimination, social responsibility and gender balance, ~~environmental friendliness and~~right to redress, sustainability, privacy and governance, should be considered high- risk from a compliance with ethical principles perspective where that is the conclusion of an impartial, regulated and external risk assessment by the national supervisory authority.

Amendment 393 #

Motion for a resolution
Annex I – part B – recital 9

(9) Any artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies~~, which entails a~~ are to be considered high -risk, of breaching the principles of safety, transparency, accountability, non-bias or non- discrimination, social responsibility and gender balance, environmental friendliness and sustainability, privacy and governance, should be considered high-risk from a compliance with ethical principles perspective where that is the conclusion of an impartial, regulated and external risk assessment by the national supervisory authorityn the basis of an impartial, objective and external risk assessment by the national supervisory authority when, considering their intended use and the critical sectors where they are employed, their autonomous operation involves a significant potential to cause harm to one or more persons, in a manner that is random and impossible to predict in advance.

Amendment 394 #

Motion for a resolution
Annex I – part B – recital 9

(9) Any artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, which entails a high risk of breaching the ~~principles of safety, transparency, accountability, non-bias or non- discrimination, social responsibility and gender balance, environ~~Union’s legal principles and fundamental fri~~endliness and sustainability, privacy and governance, should be considered high-risk from a compliance with ethical principles perspective~~ghts should be considered high-risk where that is the conclusion of an impartial, regulated and external risk assessment by the ~~national supervisory authority~~Commission, the European Parliament, the European Agency for Artificial Intelligence or a competent supervisory authority. The affected developer and deployer should be heard and its own self- assessment be taken into account for the final decision.

Amendment 395 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 9 a (new)

(9a) Determining how significant the potential to cause harm or damage by high-risk artificial intelligence, robotics and related technologies should depend on the interplay between the severity of possible harm, the likelihood that the risk materialises and the manner in which the technologies are being used. The degree of severity should be determined based on the extent of the potential harm resulting from the operation, the number of affected persons, the total value of the potential damage as well as the harm to society as a whole. The likelihood should be determined based on the role of the algorithmic calculations in the decision- making process, the complexity of the decision and the reversibility of the effects. Ultimately, the manner of usage should depend, among other things, on whether, taking into account the specific sector in which the artificial intelligence, robotics and related technologies operate, it could have legal or factual effects on important legally protected rights of the affected person, and whether the effects can reasonably be avoided.

Amendment 396 #

Motion for a resolution
Annex I – part B – recital 9 a (new)

(9a) The Commission should prepare non-binding guidelines on the methodology for compliance with this Regulation intended to developers, deployers and users. In doing so, the Commission should consult relevant stakeholders.

Amendment 397 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 10

(10) Notwithstanding the risk assessment carried out in relation to compliance with ethical principles, artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should always be assessed as to their risk on the basis of objective criteria and in line with relevant sector-specific legislation applicable in different fields such as those of health, transport, employment, justice and home affairs, media, education and culture.deleted

Amendment 398 #

Motion for a resolution
Annex I – part B – recital 10

(10) Notwithstanding the risk assessment carried out in relation to compliance with ethical principles, artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies,High-risk technologies should always be assessed as to their risk on the basis of the objective criteria ~~and in line with~~laid down in this Regulation and without prejudice to the relevant sector-specific legislation applicable in different fields such as those of health, transport, employment, justice and home affairs, media, education and culture.

Amendment 399 #

Motion for a resolution
Annex I – part B – recital 10

(10) Notwithstanding the risk assessment carried out in relation to compliance with Union law and ethical principles, artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should always be assessed as to their risk on the basis of objective criteria and in line with relevant sector-specific legislation applicable in different fields ~~such as those of health, transport, employment, justice and home affairs, media, education and culture~~.

Amendment 4 #

Motion for a resolution
Citation 7 a (new)

- having regard to the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions of 11 December 2019 on The European Green Deal,

Amendment 40 #

Motion for a resolution
Recital G

G. whereas that framework should encompass all situations requiring due consideration of the Union’s principles and values, namely development, deployment and use of the relevant high-risk technologies and their components;

Amendment 400 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 10

(10) Notwithstanding the risk assessment carried out in relation to compliance with ethical principles, artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should always be assessed as to their risk and subject to corresponding legal obligations on the basis of objective criteria and in line with relevant sector-specific legislation applicable in different fields such as those of health, transport, employment, justice and home affairs, media, education and culture.

Amendment 401 #

Motion for a resolution
Annex I – part B – recital 10 a (new)

(10a) When a high-risk technology has been considered compliant with the principles laid out in this regulation, the software, algorithms and data which are used or produced by the technology should be presumed compliant with this regulation, unless the national supervisory authority decides to conduct an assessment at its own initiative or at the request of the developer, the deployer or the user.

Amendment 402 #

Motion for a resolution
Annex I – part B – recital 10 a (new)

(10a) Any artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies considered as high-risk from a compliance with ethical principles perspective shall not be classified as low risk from a sector- specific legislation perspective.

Amendment 403 #

Motion for a resolution
Annex I – part B – recital 11

(11) Trustworthy artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies should be developed, deployed and used in a safe, transparent ~~and~~, accountable and accessible manner based on the features of robustness, resilience, security, accuracy and error identification, explainability ~~and identifiability~~, interpretability, auditability, transparency, identifiability, and access to technology, data and computing systems underlying such technologies, and in a manner that makes it possible to be ~~temporarily~~ disabled ~~and t~~or reverted to ~~historical~~a previous state, identified as restoring safe functionalities in cases of non- compliance with those ~~safety~~ features.

Amendment 404 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 11

(11) To be trustworthy high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies should be developed, deployed and used in a safe, transparent and accountable manner based on the features of robustness, resilience, security, accuracy and error identification, explainability and identifiability, and in a manner that makes it possible to be temporarily disabled and to revert to historical functionalities in cases of non- compliance with those safety features.

Amendment 405 #

Motion for a resolution
Annex I – part B – recital 12

(12) Developers, deployers and users of high-risk technologies are responsible for compliance with safety, transparency, and accountability principles to the extent of their involvement with the artificial intelligence, robotics and related technologies concerned, including the software, algorithms and data used or produced by such technologies. Developers should ensure that the technologies concerned are designed and built in line with safety features, whereas deployers and users should deploy and use the concerned technologies in full observance of those features. To this end, developers of high- risk technologies should evaluate and anticipate the risks of misuse of their own technologies, in order to respond effectively if the problem arises.

Amendment 406 #

Motion for a resolution
Annex I – part B – recital 12

(12) Developers, deployers and users arcan be responsible ~~for~~to varying degree for the compliance with safety, transparency, ~~and~~ accountability, and accessibility principles to the extent of their involvement with the artificial intelligence, robotics and related technologies concerned, including the software, algorithms and data used or produced by such technologies. Developers should ensure that the technologies concerned are designed and built in line with ~~safety~~the abovementioned features, whereas deployers and users should deploy and use the concerned technologies in full observance of those features.

Amendment 407 #

Motion for a resolution
Annex I – part B – recital 13

(13) Developers and deployers should make available to users any subsequent updates of the technologies concerned, namely in terms of software, in accordance with the obligations stipulated in the contract or laid down in Union or national law.

Amendment 408 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 13 a (new)

(13a) Where a risk assessment so indicates, developers and deployers should be required to provide for the relevant documentation on the use and design instructions, including the source code, development tools and data used by the system, made easily accessible through a mandatory legal deposit;

Amendment 409 #

Motion for a resolution
Annex I – part B – recital 14

(14) To the extent that their involvement with those technologies influences the compliance with the safety, transparency and accountability requirements set out in this Regulation, users should use artificial intelligence, robotics and related technologies in good faith. This means, in particular, that they should not use those technologies in a way that contravenes the ethical principles laid down in this legal framework and the requirements listed therein. Beyond such use in good faith, users should be exempt from any responsibility that otherwise falls upon developers and deployers as established in this Regulation.deleted

Amendment 41 #

Motion for a resolution
Recital H

H. whereas ~~a harmonised approach to~~ ethical principles relating to artificial intelligence, robotics and related technologies requires a common understanding ~~in the Union of those concepts and of concepts~~of concepts that make up the technologies, such as algorithms, and software~~, data or biometric recognition~~;

Amendment 410 #

Motion for a resolution
Annex I – part B – recital 14

(14) To the extent that their involvement with those technologies influences the compliance with the safety, transparency and accountability requirements set out in this Regulation, users should use high-risk artificial intelligence, robotics and related technologies in good faith. This means, in particular, that they should ~~not~~ use those technologies in a ~~way tha~~ccordance with the safety and use instructions provided by the developer and/or the deployer and in a way that does not contravenes the ethical principles laid down in this legal framework and the requirements listed therein. Beyond such use in good faith, users should be exempt from any responsibility that otherwise falls upon developers and deployers as established in this Regulation.

Amendment 411 #

Motion for a resolution
Annex I – part B – recital 14

(14) To the extent that their involvement with those technologies influences the compliance with the safety, transparency ~~and~~, accountability, and accessibility requirements set out in this Regulation, users should use artificial intelligence, robotics and related technologies in good faith. This means, in particular, that they should not use those technologies in a way that contravenes Union law or the ethical principles laid down in this legal framework and the requirements listed therein. Beyond such use in good faith, users should be exempt from any responsibility that otherwise falls upon developers and deployers as established in this Regulation.

Amendment 412 #

Motion for a resolution
Annex I – part B – recital 15

(15) Individuals have a right to trust technology they use to perform in a reasonable manner, to respect their trust, and protect their good faith in it. The citizens’ trust in artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, depends on the understanding and comprehension of the technical processes. The degree of explainability of such processes should depend on the context and the severity of the consequences of an erroneous or inaccurate output of those technical processes, and needs to be sufficient for challenging them and seeking redress. Auditability ~~and~~, traceability, and transparency should remedy the possible unintelligibility of such technologies.

Amendment 413 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 15

(15) The citizens’ trust in high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, depends on the understanding and comprehension of the technical processes. The degree of explainability of such processes should depend on the context and the severity of the consequences of an erroneous or inaccurate output of those technical processes, and needs to be sufficient for challenging them and seeking redress. Auditability and traceability should remedy the possible unintelligibility of such technologies.

Amendment 414 #

Motion for a resolution
Annex I – part B – recital 16

(16) Society’s trust in ~~artificial intelligence, robotics and related~~ high-risk technologies, including the software, algorithms and data used or produced by such technologies, depends on the degree to which their assessment, auditability and traceability are enabled in the technologies concerned. Where the extent of their involvement so requires, developers should ensure that such technologies are designed and built in a manner that enables such an assessment, auditing and traceability. DWithin the limits of what is technically possible, developers, deployers and users should ensure that artificial intelligence, robotics and related technologies are deployed and used in full respect of transparency requirements, and allowing auditing and traceability.

Amendment 415 #

Motion for a resolution
Annex I – part B – recital 16

(16) Society’s trust in artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, depends on the degree to which their assessment, auditability and traceability are enabled in the technologies concerned. Where the extent of their involvement so requires, developers should ensure that such technologies are designed and built in a manner that enables such an assessment, auditing and traceability. Deployers and users should ensure that artificial intelligence, robotics and related technologies are deployed and used in full respect of accessibility, and transparency requirements, and allowing auditing and traceability.

Amendment 416 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 16

(16) Society’s trust in high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, depends on the degree to which their assessment, auditability and traceability are enabled in the technologies concerned. Where the extent of their involvement so requires, developers should ensure that such technologies are designed and built in a manner that enables such an assessment, auditing and traceability. Deployers and users should ensure that artificial intelligence, robotics and related technologies are deployed and used in full respect of transparency requirements, and allowing auditing and traceability.

Amendment 417 #

Motion for a resolution
Annex I – part B – recital 16 a (new)

(16a) In order to ensure transparency and accountability, citizens should be informed when a system uses artificial intelligence, when AI systems personalise a product or service to its users, whether they can switch off or restrain the personalisation and when they are faced with an automated-decision making technology. Furthermore, transparency measures should be accompanied, as far as this is technically possible, by clear and understandable explanations of the data used, of the algorithm, of its purpose, of its outcomes, and of its potential dangers;

Amendment 418 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 17

(17) Bias in and discrimination by software, algorithms and data is unlawful and should be addressed by regulating the processes through which they are designed and usdeployed.

Amendment 419 #

Motion for a resolution
Annex I – part B – recital 18

(18) Software, algorithms and data ~~used or produc~~produced or destined to be used by artificial intelligence, robotics and related technologies should be considered biased where~~, for example,~~ - among others - they display suboptimal results in relation to any person or group of persons, on the basis of a prejudiced personal, social or partial perception and subsequent processing of data relating to their traits.

Amendment 42 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital H

H. whereas a harmonised approach to ethical principles relating to high-risk artificial intelligence, robotics and related technologies requires a common understanding in the Union of those concepts and of concepts such as algorithms, software, data or biometric recognition;

Amendment 420 #

Motion for a resolution
Annex I – part B – recital 18

(18) Software, algorithms and data used or produced by ~~artificial intelligence, robotics and related~~high-risk technologies1a should be considered biased where, for example, they display suboptimal results in relation to any person or group of persons, on the basis of a prejudiced personal, social or partial perception and subsequent processing of data relating to their traits. __________________ 1aFrom this point, “artificial intelligence, robotics and related technologies” should be replaced by “high-risk technologies” throughout the recitals.

Amendment 421 #

Motion for a resolution
Annex I – part B – recital 19

(19) In line with Union law, software, algorithms and data ~~used or produc~~produced or destined to be used by artificial intelligence, robotics and related technologies should be considered discriminatory where they treat a person or group of persons differently, including by putting them at a disadvantage when compared to others, based on grounds such as their personal traits, without objective or reasonable justificationproduce outcomes that disadvantage a person based on the basis of a group characteristic that is considered to be irrelevant or otherwise unacceptable or where – despite apparent neutrality of the technologies - such outcomes have disproportionate negative effects on people belonging to such groups, unless objectively justified by a legitimate aim, and the means of achieving that aim are appropriate and necessary.

Amendment 422 #

Motion for a resolution
Annex I – part B – recital 20

(20) In line with Union law, legitimate aims that might objectively justify any differential treatment between persons or group of persons are the protection of public safety, security and health, the prevention of criminal offences, the protection of individual rights and freedoms, fair representation and objective professional requirements.deleted

Amendment 423 #

Brando BENIFEI

Motion for a resolution
Annex I – part B – recital 20

(20) In line with Union law, legitimate aims that might objectively justify any differential treatment between persons or group of persons are the protection of public safety, security and health, the pr~~even~~otection of ~~criminal offences, the protection of individu~~fundamental rights and freedoms, fair representation and objective professional requirements.

Amendment 424 #

Motion for a resolution
Annex I – part B – recital 21

(21) Artificial intelligence, robotics and related technologies, including software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, should perform on the basis of sustainable progress. Such technologies should not run count~~ribute comprehensively to~~er to the interests of the people and the planet, the cause of preservation of the environment, the green transition, and the achievement of the Sustainable Development Goals outlined by the United Nations with a view to enabling future generations to flourish. Such technologies can support the monitoring of adequate progress on the basis of sustainability and social cohesion indicators, and by using responsible research and innovation tools requiring the mobilisation of resources by the Union and its Member States to support and invest in projects addressing those goals.

Amendment 425 #

Motion for a resolution
Annex I – part B – recital 21

(21) Artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, should ~~perform on the basis of~~contribute to sustainable progress. Such technologies ~~should contribute comprehensively to the~~can also play an important role in achiev~~ement~~ing of the Sustainable Development Goals outlined by the United Nations with a view to enabling future generations to flourish. Such technologies can support the monitoring of adequate progress on the basis of sustainability and social cohesion indicators, ~~and~~ by using responsible research and innovation tools requiring the mobilisation of resources by the Union and its Member States to support and invest in projects addressing those goals.

Amendment 426 #

Motion for a resolution
Annex I – part B – recital 22

(22) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, should in no way ~~cause~~purposefully cause or consciously accept by design injury or harm of any kind to individuals or society. Accordingly, such technologies should be developed, deployed and used in a socially responsible manner.

Amendment 427 #

Motion for a resolution
Annex I – part B – recital 22

(22) The development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should in no way cause injury or harm of any kind to individuals or society. Accordingly, such technologies should be developed, deployed and used in a socially responsible manner. and, therefore,

Amendment 428 #

Motion for a resolution
Annex I – part B – recital 23

(23) ~~For the purposes of this Regulation, d~~Developers, deployers and users should be held responsible, to the extent of their involvement in the artificial intelligence, robotics and related technologies concerned, for any injury or harm inflicted upon individuals and society in accordance with Union and national liability rules.

Amendment 429 #

Motion for a resolution
Annex I – part B – recital 24

(24) In particular, the developers who take decisions that determine and control the course or manner of the development of artificial intelligence, robotics and related technologies, as well as the deployers who are involved in their deployment with an operating or managing function, should be generally considered responsible for avoiding the occurrence of any such injury or harm, by putting adequate measures in place during the development process and thoroughly respecting such measures during the deployment phase, respectively.deleted

Amendment 43 #

Motion for a resolution
Recital I

I. whereas action at Union level is justified by the need to avoid regulatory fragmentation or a series of national regulatory provisions with no common denominator for a homogenous application of common ethical principles when developing, deploying and using artificial intelligence, robotics and related technologies;

Amendment 430 #

Motion for a resolution
Annex I – part B – recital 25

(25) Socially responsible artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, can be defined as technologies which ~~both~~contribute to find solutions that safeguard and promote a number of different aspects of society, most notably democracy, health and economic prosperity, equality of opportunity, workers’ and social rights, diverse and independent media and objective and freely available information, allowing for public debate, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity. Nevertheless, those requirements shall be applicable only to high-risk artificial intelligence, robotics and related technologies. They are also those that are developed, deployed and used having due regard for their ultimate impact on the physical and mental well- being of citizens.

Amendment 431 #

Motion for a resolution
Annex I – part B – recital 25

(25) Socially responsible artificial intelligence, robotics and related technologies, including the software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, can be defined as technologies which both safeguard and promote a number of different aspects of society, most notably democracy, health and economic prosperity, equality of opportunity, workers’ and social rights, diverse and independent media and objective and freely available information, allowing for public debate, quality education, cultural and linguistic diversity, gender balance, digital literacy, innovation and creativity. They are also those that are developed, deployed and used having due regard for their ultimate impact on the physical and mental well-being of citizens.

Amendment 432 #

Motion for a resolution
Annex I – part B – recital 26

(26) These technologies should also be developed, deployed and used with a view to supporting social inclusion, plurality, solidarity, fairness, equality and cooperation and their potential in that context should be maximized and explored through research and innovation projects. The Union and its Member States should ~~therefore~~ mobilise their resources for the purpose of supporting and investing in such projects.

Amendment 433 #

Motion for a resolution
Annex I – part B – recital 26

(26) These technologies ~~should~~can also be developed, deployed and used with a view to supporting social inclusion, plurality, solidarity, fairness, equality and cooperation and their potential in that context should be maximized and explored through research and innovation projects. The Union and its Member States should therefore mobilise their resources for the purpose of supporting and investing in such projects.

Amendment 434 #

Motion for a resolution
Annex I – part B – recital 27

(27) Projects relating to the potential of artificial intelligence, robotics and related technologies to deal with the question of social well-being should be carried out on the basis of responsible research and innovation tools so as to guarantee the compliance with ethical principles of those projects from the outset.deleted

Amendment 435 #

Motion for a resolution
Annex I – part B – recital 28

(28) The development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, should take into consideration their environmental footprint and should not cause harm to the environment during their lifecycle and across their entire supply chain. Accordingly, such technologies should be developed, deployed and used in a~~n environmentally friendly manner that~~ manner that preserves the environment, minimises and remedies their environmental footprint and contributes to the green transition, supports the achievement of climate neutrality and circular economy goals.

Amendment 436 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 28

(28) The development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should take into consideration their environmental footprint ~~and should not cause harm to the environment during their lifecycle and across their entire supply chain. Accordingly,~~. In line with the obligations laid down in Union law, high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies should be developed, deployed and used in an environmentally ~~friendly~~sustainable manner that supports the achievement of climate neutrality and circular economy goals.

Amendment 437 #

Motion for a resolution
Annex I – part B – recital 28

(28) TWhere applicable, the development, deployment and use of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should take into consideration their environmental footprint and should not cause harm to the environment during their lifecycle and across their entire supply chain. Accordingly, such technologies should be developed, deployed and used in an environmentally friendly manner that supports the achievement of climate neutrality and circular economy goals.

Amendment 438 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 29

(29) For the purposes of this Regulation, developers, deployers and users should be held responsible, to the extent of their involvement in the development, deployment or use of the artificial intelligence, robotics and related technologies concerned, for any harm caused to the environment.deleted

Amendment 439 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Annex I – part B – recital 29

(29) ~~For the purpo~~Developers, deployers and users of ~~this Regulation, developers, deployers and users~~high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, should be held responsible, to the extent of their involvement in the development, deployment or use of the artificial intelligence, robotics and related technologies concerned, for any harm caused to the environment in accordance with the applicable environmental liability rules.

Amendment 44 #

Motion for a resolution
Recital I

I. whereas action at Union level is justified by the need for a homogenous application of common ethical principles when developing, deploying and using artificial intelligence, robotics and related technologies; whereas clear rules are needed where major risks are at stake;

Amendment 440 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 30

(30) In particular, the developers who take decisions that determine and control the course or manner of the development of artificial intelligence, robotics and related technologies, as well the deployers who are involved in their deployment with an operating or managing function, should be generally considered responsible for avoiding the occurrence of such harm, namely by respectively putting adequate measures in place during the development process and thoroughly respecting such measures during the deployment phase.deleted

Amendment 441 #

Motion for a resolution
Annex I – part B – recital 30

Amendment 442 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 31

(31) These technologies should also be developed, deployed and used with a view to supporting the achievement of environmental goals prescribed in Union law, such as reducing waste production, diminishing the carbon footprint, preventing climate change and avoiding environmental degradation, and their potential in that context should be maximized and explored through research and innovation projects. The Union and the Member States should therefore mobilise their resources for the purpose of supporting and investing in such projects.

Amendment 443 #

Motion for a resolution
Annex I – part B – recital 31

(31) TWhere applicable, these technologies should also be developed, deployed and used with a view to supporting the achievement of environmental goals such as reducing waste production, diminishing the carbon footprint, preventing climate change and avoiding environmental degradation, and their potential in that context should be maximized and explored through research and innovation projects. The Union and the Member States should therefore mobilise their resources for the purpose of supporting and investing in such projects.

Amendment 444 #

Motion for a resolution
Annex I – part B – recital 31

(31) These technologies should also be developed, deployed and used with a view to supporting the achievement of environmental goals such as ~~reduc~~ending waste production, diminishing the carbon footprint, pre~~vent~~medying climate change and ~~avoiding environmental degradation~~preserving the environment, and their potential in that context should be maximized and explored through research and innovation projects. The Union and the Member States should therefore mobilise their resources for the purpose of supporting and investing in such projects.

Amendment 445 #

Motion for a resolution
Annex I – part B – recital 33

(33) Any artificial intelligence, robotics and related technologies, including software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, developed, deployed and used in the Union should fully respect Union citizens’ rights to privacy and protection of personal data. In particular, their development, deployment and use should be in accordance with Regulation (EU) 2016/679 of the European Parliament and of the Council1 and Directive 2002/58/EC of the European Parliament and of the Council2 . __________________ 1Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1). 2 Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications) (OJ L 201, 31.7.2002, p. 37).

Amendment 446 #

Motion for a resolution
Annex I – part B – recital 33

(33) ~~Any~~High-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed and used in the Union should fully respect Union citizens’ rights to privacy and protection of personal data. In particular, their development, deployment and use should be in accordance with Regulation (EU) 2016/679 of the European Parliament and of the Council1 and Directive 2002/58/EC of the European Parliament and of the Council2 . __________________ 1Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1). 2 Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications) (OJ L 201, 31.7.2002, p. 37).

Amendment 447 #

Motion for a resolution
Annex I – part B – recital 34

(34) TIn particular, the ethical boundaries of the use of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, should be duly considered when using remote recognition technologies, such as biometric recognition, to automatically identify individuals. When these technologies are used by public authorities during times of national emergency, such as during a national health crisis, the use should be proportionate and criteria for that use defined in order to be able to determine whether, when and how it should take place, and such use should be mindful of its psychological and sociocultural impact with due regard for human dignity and the fundamental rights set out in the Charter.

Amendment 448 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 34

(34) The ethical boundaries of the use of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, should be duly considered when using remote recognition technologies, such as ~~biometric~~recognition of biometric features, notably facial recognition, to automatically identify individuals. When these technologies are used by public authorities ~~during times of national emergency, such as during a national health crisis, the use should be proportionate and criteria for that use defined in order to be able~~, the use should always be voluntary, proportionate, targeted and limited to specific objectives, restricted in time and carried out in accordance with Union law, with due regard to human rights and dignity, autonomy and self-determination of the individual, and the fundamental rights set out in the Charter. Criteria for and limits to that use should be subjected to judicial review and submitted to democratic scrutiny and debate involving civil society, in order to determine whether, when and how it should take place, and such use should be mindful of its psychological and sociocultural impact ~~with due regard for human dignity and the fundamental rights set out in the Charter~~.

Amendment 449 #

Motion for a resolution
Annex I – part B – recital 34

Amendment 45 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital I

I. whereas action at Union level is justified ~~by the need for~~to ensure a homogenous application of common ethical principles enshrined in law when developing, deploying and using artificial intelligence, robotics and related technologies;

Amendment 450 #

Motion for a resolution
Annex I – part B – recital 35

(35) Governance that is based on relevant standards enhances safety and promotes the increase of citizens’ trust in the development, deployment and use of artificial intelligence, robotics and related technologies including software, algorithms and data used or produced by such technologies.deleted

Amendment 451 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 35

(35) Governance that is based on relevant standards enhances safety and promotes the increase of citizens’ trust in the development, deployment and use of high-risk artificial intelligence, robotics and related technologies including software, algorithms and data used or produced by such technologies.

Amendment 452 #

Motion for a resolution
Annex I – part B – recital 35 a (new)

(35a) Public authorities should conduct fundamental rights impact assessment before deploying high-risk technologies which replace public power decisions and which have a direct and significant impact on citizen’s rights and obligations. In addition, these technologies should allow for human verification and due process, especially in the areas of justice and law enforcement, where fundamental rights protected by the Charter, are at stake.

Amendment 453 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Karen MELCHIOR

Motion for a resolution
Annex I – part B – recital 36

(36) ~~Among~~Developers, deployers and users should continue to observe the existing relevant governance standards are, for example, the ‘Ethics Guidelines for Trustworthy AI’ drafted by the High-Level Expert Group on Artificial Intelligence set up by the European Commission, and other technical standards adopted by the European Committee for Standardization (CEN), the European Committee for Electrotechnical Standardization (CENELEC), and the European Telecommunications Standards Institute (ETSI), at European level, the International Organization for Standardization (ISO) and the Institute of Electrical and Electronics Engineers (IEEE), at international level.

Amendment 454 #

Motion for a resolution
Annex I – part B – recital 36

(36) Among the existing relevant governance standards are, for example, the ‘Ethics Guidelines for Trustworthy AI’ drafted by the High-Level Expert Group on Artificial Intelligence set up by the European Commission, and ~~other~~ technical standards such as those adopted by the European Committee for Standardization (CEN), the European Committee for Electrotechnical Standardization (CENELEC), and the European Telecommunications Standards Institute (ETSI), at European level, the International Organization for Standardization (ISO) and the Institute of Electrical and Electronics Engineers (IEEE), at international level.

Amendment 455 #

Motion for a resolution
Annex I – part B – recital 37

(37) Sharing and use of data by multiple participants is sensitive and therefore the development, deployment and use of artificial intelligence, robotics and related technologies should be governed by relevant standards and protocols reflecting the requirements of quality, integrity, security, privacy and control. The data governance strategy should focus on the processing, sharing of and access to such data, including its proper management and traceability, and guarantee the adequate protection of data belonging to vulnerable groups, including people with disabilities, patients, children, minorities and migrants. or other groups at risk of exclusion;

Amendment 457 #

Motion for a resolution
Annex I – part B – recital 37

(37) Sharing and use of data by multiple participants is sensitive and therefore the development, deployment and use of high- risk artificial intelligence, robotics and related technologies should be governed by relevant rules, standards and protocols reflecting the requirements of quality, integrity, security, privacy and control. The data governance strategy should focus on the processing, sharing of and access to such data, including its proper management and traceability, and guarantee the adequate protection of data belonging to vulnerable groups, including people with disabilities, patients, children, minorities and migrants.

Amendment 458 #

Esteban GONZÁLEZ PONS

Motion for a resolution
Annex I – part B – recital 37

(37) Sharing and use of data by multiple participants is sensitive and therefore the development, deployment and use of artificial intelligence, robotics and related technologies should be governed by relevant standards and protocols reflecting the requirements of quality, integrity, security, reliability, privacy and control. The data governance strategy should focus on the processing, sharing of and access to such data, including its proper management and traceability, and guarantee the adequate protection of data belonging to vulnerable groups, including people with disabilities, patients, children, minorities and migrants.

Amendment 459 #

Motion for a resolution
Annex I – part B – recital 37

(37) Sharing and use of data by multiple participants is sensitive and therefore the development, deployment and use of artificial intelligence, robotics and related technologies should be governed by relevant standards and protocols reflecting the requirements of quality, integrity, security, privacy and control. The data governance strategy should focus on the processing, sharing of and access to such data, including its proper management, auditability and traceability, and guarantee the adequate protection of data belonging to vulnerable groups, including people with disabilities, patients, children, minorities and migrants.

Amendment 46 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital I

I. whereas action at Union level is justified by the need for a homogenous application of common ethical principles when developing, deploying and using high-risk artificial intelligence, robotics and related technologies;

Amendment 460 #

Motion for a resolution
Annex I – part B – recital 38

(38) ~~The effective application of the ethical principles laid down in this Regulation will largely depend on~~ Member States’ should appoint~~ment of an~~ independent public authority to act as a supervisory authority. ~~In particular, e~~Each national supervisory authority should be responsible for assessing ~~and monitoring~~ the compliance of ~~artificial intelligence, robotics and related technologies considered a high-risk in light of the obligations set out in~~high-risk technologies with this Rregulation.

Amendment 461 #

Motion for a resolution
Annex I – part B – recital 38

(38) The effective application of the ethical principles laid down in this Regulation will largely depend on Member States’ appointment of an independent public authority, or an authority with a public purpose to act as a supervisory authority. In particular, each ~~national~~competent supervisory authority should be responsible for assessing and monitoring the compliance of artificial intelligence, robotics and related technologies considered a high-risk in light of the obligations set out in this Regulation.

Amendment 462 #

Motion for a resolution
Annex I – part B – recital 38

(38) The effective application of the ethical principles laid down in this Regulation will largely depend on Member States’ appointment of an independent public authority to act as a supervisory authority. In particular, each national supervisory authority should be responsible for ~~assessing and monitoring the compliance of~~identifying artificial intelligence, robotics and related technologies considered a high-risk and for assessing and monitoring the compliance of these technologies in light of the obligations set out in this Regulation.

Amendment 463 #

Motion for a resolution
Annex I – part B – recital 38

(38) The effective application of the ethical principles laid down in this Regulation will largely depend on Member States’ appointment of an independent public authority to act as a supervisory authority. In particular, each national supervisory authority should be responsible for assessing and monitoring the compliance of artificial intelligence, robotics and related technologies considered a high-risk in light of the ob~~ligations~~jective criteria set out in this Regulation.

Amendment 464 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 39

(39) Each national supervisory authority shall also carry the responsibility of ~~regulating~~ the governance of these technologies in accordance with a common framework prepared by the European Commission or another relevant institution, body, office or agency of the Union designated for this task. They therefore have an important role to play in promoting the trust and safety of Union citizens, as well as in enabling a democratic, pluralistic and equitable society.

Amendment 465 #

Motion for a resolution
Annex I – part B – recital 39

(39) Each national supervisory authority ~~shall also carry the responsibility of regulating the governance of these technologies. They therefore~~will have an important role to play in promoting the trust and safety of Union citizens, as well as in enabling a democratic, pluralistic and equitable society.

Amendment 466 #

Motion for a resolution
Annex I – part B – recital 39

(39) Each ~~national~~competent supervisory authority shall also carry the responsibility of regulating the governance of these technologies. They therefore have an important role to play in promoting the trust and safety of Union citizens, as well as in enabling a democratic, pluralistic and equitable society.

Amendment 467 #

Motion for a resolution
Annex I – part B – recital 39 a (new)

(39a) National supervisory authorities shall cooperate with the authorities responsible for assessing and monitoring artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies and enforcing their compliance in the light of sectorial legislation in order to identify technologies which are high-risk from an ethical perspective and in order to supervise the implementation of required and appropriate measures where such technologies are identified;

Amendment 468 #

Motion for a resolution
Annex I – part B – recital 40

(40) ~~National~~Competent supervisory authorities should engage in substantial and regular cooperation with each other, as well as with the European Commission and other relevant institutions, bodies, offices and agencies of the Union, in order to guarantee a coherent cross-border action, and allow for consistent development, deployment and use of these technologies within the Union in compliance with the ethical principles laid down in this Regulation.

Amendment 469 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 40 a (new)

(40a) In the context of such cooperation, national supervisory authorities, together with the European Commission and other relevant institutions, bodies, offices and agencies of the Union, should elaborate a common and exhaustive list of high-risk artificial intelligence, robotics and related technologies in line with the criteria set out in this Regulation and develop common criteria and an application process should be developed for the granting of a European certificate of ethical compliance.

Amendment 47 #

Motion for a resolution
Recital I a (new)

Ia. whereas the European Union needs to recognise, harness and promote the benefits of artificial intelligence, robotics and related technologies for the society, while democratically deciding on the limitations to be laid down and safeguards to be provided to ensure the development, deployment and use of ethically embedded technologies that respect the Charter of Fundamental Rights of the European Union;

Amendment 470 #

Motion for a resolution
Annex I – part B – recital 41

(41) ~~National~~Competent supervisory authorities should ensure the gathering of a maximum number of stakeholders such as industry, businesses, social partners, researchers, consumers and civil society organisations, and provide a pluralistic forum for reflection and exchange of views, to facilitate cooperation and collaboration between stakeholders, in particular from academia, research, industry, civil society and individual experts, so as to achieve comprehensible and accurate conclusions for the purpose of guiding how governance is regulated.

Amendment 471 #

Motion for a resolution
Annex I – part B – recital 42

(42) Additionally, these ~~national~~competent supervisory authorities should provide professional administrative guidance and support to developers, deployers and users, particularly small and medium-sized enterprises or start-ups, encountering challenges as regards complying with the principles laid down in this Regulation.

Amendment 472 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 42 a (new)

(42a) Full harmonisation approach is needed at the European level. Therefore, the European Commission shall be tasked to find an appropriate solution to structure such an approach. The goal is to avoid creation of another one agency. Instead, we have to find strict rules and guidelines for cooperation between Member States.

Amendment 473 #

Motion for a resolution
Annex I – part B – recital 42 a (new)

(42a) The Commission should establish binding guidelines to be followed by the national supervisory authorities when conducting their compliance assessment.

Amendment 474 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 43

(43) Whistle-blowing brings potential and actual breaches of Union law to the attention of authorities with a view to preventing injury, harm or damage that would otherwise occur. In addition, reporting procedures ameliorate the information flow within companies and organisations, thus mitigating the risk of flawed or erroneous products or services being developed. Companies and organisations developing, deploying or using artificial intelligence, robotics and related technologies, including data used or produced by those technologies, should set up reporting channels and persons reporting breaches should be protected from retaliation.deleted

Amendment 475 #

Motion for a resolution
Annex I – part B – recital 44

(44) The rapid development of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, as well as of the technical machine learning, reasoning processes and other technologies underlying that development are unpredictable. As such, it is both appropriate and necessary to establish a review mechanism in accordance with which, in addition to its reporting on the application of the Regulation, the Commission is to regularly submit a report concerning the possible modification of the scope of application of this Regulation. In addition, the Commission should review, if necessary, every six months the annex of this Regulation by means of delegated act.

Amendment 476 #

Motion for a resolution
Annex I – part B – recital 45

(45) Since the objective of this Regulation, namely to establish a common legal framework of ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies in the Union, cannot be sufficiently achieved by the Member States, but can rather, by reason of its scale and effects, be better achieved at Union level, the Union may adopt measures, in accordance with the principle of subsidiarity as set out in Article 5 of the Treaty on European Union. In accordance with the principle of proportionality, as set out in that Article, this Regulation does not go beyond what is necessary in order to achieve that objective.

Amendment 477 #

Motion for a resolution
Annex I – part B – recital 46

(46) Action at Union level as set out in this Regulation would be best achieved through the establishment of a European Agency for Artificial Intelligence. Such a body would be essential in coordinating the mandates and actions of the national supervisory authorities in each Member State, outlining objective criteria for the risk assessment of artificial intelligence, robotics and related technologies, developing and issuing a certification of compliance with the ethical principles laid down in this Regulation, supporting regular exchanges with concerned stakeholders and civil society, promoting the Union’s approach through international cooperation and ensuring a consistent reply worldwide to the opportunities and risks inherent in these technologies.deleted

Amendment 478 #

Motion for a resolution
Annex I – part B – recital 46

Amendment 479 #

Motion for a resolution
Annex I – part B – recital 46

Amendment 48 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital J

J. whereas common ethical principles are only efficient where they are also enshrined in law, and those responsible for ensuring, assessing and monitoring compliance are identified;

Amendment 480 #

Motion for a resolution
Annex I – part B – recital 46

Amendment 481 #

Motion for a resolution
Annex I – part B – recital 46

Amendment 482 #

Motion for a resolution
Annex I – part B – recital 46

(46) Action at Union level as set out in this Regulation would be best achieved through the establishment of a European Agency for Artificial Intelligence. Such a body would be essential in coordinating the mandates and actions of the ~~national~~competent supervisory authorities in ~~each~~ Member States, outlining objective criteria for the risk assessment of artificial intelligence, robotics and related technologies, developing and issuing a certification of compliance with the ethical principles laid down in this Regulation, supporting regular exchanges with concerned stakeholders and civil society, creating a centre of expertise, bringing together academia, research, industry, and individual experts at Union level to foster exchange of knowledge and technical expertise, and promoting the Union’s approach through international cooperation and ensuring a consistent reply worldwide to the opportunities and risks inherent in these technologies.

Amendment 483 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – recital 46

(46) Action at Union level as set out in this Regulation would be best achieved through the establishment of a European Agency for Artificial Intelligence. Such a body would be essential in coordinating the mandates and actions of the national supervisory authorities in each Member State, outlining objective criteria for the risk assessment of artificial intelligence, robotics and related technologies, drafting a common framework for the governance of the development, deployment and use of these technologies, developing and issuing a certification of compliance with the ethical principles laid down in this Regulation, supporting regular exchanges with concerned stakeholders and civil society, promoting the Union’s approach through international cooperation and ensuring a consistent reply worldwide to the opportunities and risks inherent in these technologies.

Amendment 484 #

Motion for a resolution
Annex I – part B – before Article 1 – chapter title (new)

Chapter I: General provisions

Amendment 485 #

Motion for a resolution
Annex I – part B – Article 1 – paragraph 1

The purpose of this Regulation is to establish a comprehensive and forward- looking regulatory framework of ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies in the Union.

Amendment 486 #

Motion for a resolution
Annex I – part B – Article 1 – paragraph 1

The purpose of this Regulation is to establish a European common legal regulatory framework of ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies in the Union.

Amendment 487 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 1 – paragraph 1

The purpose of this Regulation is to establish a regulatory framework of ethical principles for the development, deployment and use of high-risk artificial intelligence, robotics and related technologies in the Union.

Amendment 488 #

Motion for a resolution
Annex I – part B – Article 2 – paragraph 1

1. This Regulation applies to high- risk artificial intelligence, robotics and related technologies, hereafter “high-risk technologies”1a including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union. 2. Within the meaning of this Regulation are considered being high- risk technologies, technologies that: (a) are used in sectors, where given the characteristics of the activities typically undertaken, significant risks can be expected to occur from the viewpoint of protection of safety and fundamental rights and freedoms, and (b) are used in such a manner, that such risks can be expected to occur from the viewpoint of protection of safety and fundamental rights and freedoms 3. The high-risk technologies mentioned in Paragraph 2 shall be listed in an Annex to this Regulation. The Commission is empowered to adopt delegated acts in accordance with Article 16(a), to amend the exhaustive list, by: (a) including new types of high-risk artificial intelligence, robotics and related technologies; (b) deleting types of technologies that can no longer be considered to pose a high-risk; (c) changing the critical sectors for existing high-risk technologies. Any delegated act amending the Annex shall come into force six months after its adoption. When determining new critical sectors and/or high-risk technologies to be inserted by means of delegated acts in the Annex, the Commission shall take full account of the criteria set out in this Regulation, in particular those set out in paragraph 2 of this Article. __________________ 1aFrom this point “artificial intelligence, robotics and related technologies” is replaced by “high-risk technologies” throughout the Regulation

Amendment 489 #

Motion for a resolution
Annex I – part B – Article 2 – paragraph 1

This Regulation applies to artificial intelligence, robotics and related technologies, including software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, developed, deployed or used in the Union.

Amendment 49 #

Motion for a resolution
Recital J a (new)

Ja. whereas there are serious concerns that the current EU legal framework, including the consumer law acquis, data protection legislation, product safety and market surveillance legislation, as well as antidiscrimination legislation is not fit for purpose to effectively tackle the risks created by artificial intelligence, robotics and related technologies;

Amendment 490 #

Motion for a resolution
Annex I – part B – Article 2 – paragraph 1

This Regulation applies to high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union.

Amendment 491 #

Motion for a resolution
Annex I – part B – Article 2 – paragraph 1 a (new)

1a. This Regulation shall not apply to artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union which are not considered high-risk.

Amendment 492 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a

(a) ‘artificial intelligence systems’ means ~~software~~ systems that~~, inter alia, collect, process and interpret structured or unstructured data, identify patterns and establish models in order to reach conclusions or take~~ display intelligent behaviour by analysing certain input and taking action, with some degree of autonomy, to achieve specific goals. AI- systems can be purely software-based, acti~~ons~~ng in the ~~physic~~virtual wor ~~virtual dimension bas~~ld, or can be embedded oin ~~such conclusion~~hardware devices;

Amendment 493 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a

(a) ‘artificial intelligence’ means ~~software systems that, inter alia, collect, process and interpret structured or unstructured data,~~methods and procedures that enable technical systems to perceive their environment, process what is perceived and solve problems inde~~ntify patterns and establish models in order to reach conclusions or take actions in the physical or virtual dimension based on such conclus~~pendently, make decisions, act and learn from the consequences of those decisions and actions;

Amendment 494 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a

(a) ‘artificial intelligence’ means ~~software~~a systems that~~, inter alia, collect, process and interpret structured or unstructured data, identify patterns and establish models in order to reach conclusions or take~~ displays intelligent behaviour by analysing their environment and taking actions, with some degree of autonomy, to achieve specific goals; AI- systems can be purely software-based, acti~~ons~~ng in the ~~physical or virtual dimension bas~~virtual world, or can be embedded oin ~~such conclusion~~hardware devices;

Amendment 495 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a

(a) ‘artificial intelligence’ means software systems that, inter alia, collect, process and interpret structured or unstructured data, identify patterns and establish models in order to reach conclusions or take actions in the physical or virtual dimension basdisplay intelligent behaviour by analysing certain input and taking action, with some degree of autonomy, to achieve specific goals. AI systems can be purely software-based, acting in virtual world, or can be embedded oin ~~such conclusion~~hardware devices;

Amendment 496 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a

(a) ‘artificial intelligence’ means ~~software~~ systems that display intelligent behaviour by, inter alia, collecting, process ~~and interpret structured or unstructured data, identify patterns and establish models in order~~ing, analysing, and interpreting their environment and taking actions, with some degree of autonomy, to reach ~~conclusions or take ac~~ieve specific goals1a; __________________ 1aDefinition as in the ~~physical or virtual dimension based on such conclusions;~~European Commission Communication COM(2018) 237 final, 25.04.2018, page 1, adapted.

Amendment 497 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a a (new)

(aa) ‘automated decision making’, decision support, or decision informing systems are procedures in which decisions are initially, partly or completely, delegated to another person or corporate entity by way of using a software or a service, who then in turn use automatically executed decision-making models to perform an action;

Amendment 498 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point b

(b) ‘robotics’ means ~~technologies that enabl~~automatically controlled, reprogrammable, multi- purpose machines1a to perform ~~tasks traditionally performed by human beings~~actions in the physical world including by way of artificial intelligence or related technologies; __________________ 1aFrom the definition for industrial robots in ISO 8373.

Amendment 499 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point b

(b) ‘robotics’ means technologies that enable machines to perform tasks traditionally performed or initiated by human beings including by way of artificial intelligence or related technologies;

Amendment 5 #

Motion for a resolution
Citation 7 a (new)

- having regard to the Charter of Fundamental Rights of the European Union,

Amendment 50 #

Motion for a resolution
Recital J b (new)

Jb. whereas ethical guidance, such as the principles adopted by the High-Level Expert Group on Artificial Intelligence, provides a good starting point but is not enough to ensure that businesses act fairly and guarantee the effective protection of individuals;

Amendment 500 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point c

(c) ‘related technologies’ means technologies that enable software to control with a partial or full degree of autonomy a physical or virtual process, technologies capable of detecting the identity of persons or specific features of persons by way of their biometric data, and technologies that copy or otherwise make use of human traits;deleted

Amendment 501 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point d

~~(d) ‘software’ means a set of instructions that are expressed in code and necessary for a computer to operate and to execute tasks;~~deleted

Amendment 502 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point e

~~(e) ‘algorithms’ means a model for calculations or other problem-solving operations carried out by software when executing a task;~~deleted

Amendment 503 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point f

~~(f) ‘data’ means information defined as and stored in code;~~deleted

Amendment 504 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point f a (new)

(fa) ‘autonomous’ means an artificial intelligence, robotics or related technology that operates by perceiving certain input and without needing to follow a set of pre-determined instructions, despite its behaviour being constrained by the goal it was given and other relevant design choices made by its developer;

Amendment 505 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point f b (new)

(fb) ‘high risk’ means a significant potential in an autonomously operating artificial intelligence, robotics and related technology to cause harm or damage to one or more persons in a manner that is random and impossible to predict in advance, considering its intended use and the critical sector where it is employed;[FDCM1] the significance of the potential depends on the interplay between the severity of possible harm or damage, the likelihood that the risk materialises and the manner in which the AI-system is being used;

Amendment 506 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point g

(g) ‘development’ means the construction and design of algorithms, the writing and design of software or the collection, storing and management of data for the purpose of creating or training artificial intelligence, robotics and related technologies or for the purpose of creating a new application for existing artificial intelligence, robotics and related technologies;deleted

Amendment 507 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point h

(h) ‘developer’ means any natural or legal person who takes decisions that determine and control the course or manner of the development of artificial intelligence, robotics and related technologies;deleted

Amendment 508 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point i

(i) ‘deployment’ means the operation and management of artificial intelligence, robotics and related technologies, as well as their placement on the market or otherwise making them available to users;deleted

Amendment 509 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point j

~~(j) ‘deployer’ means any natural or legal person who is involved in the deployment of artificial intelligence, robotics and related technologies, and has an operating or managing function;~~deleted

Amendment 51 #

Motion for a resolution
Recital K

K. whereas each Member State should establish a national supervisory authority responsible for ensuring, assessing and monitoring the compliance~~, and for enabling~~ of the development, deployment and use of artificial intelligence, robotics and related technologies with these ethical principles; and to allow discussions and exchange ~~of point~~s of views in close cooperation with ~~the concerned~~ stakeholders and ~~the~~ civil society;

Amendment 510 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point j

(j) ‘deployer’ means any natural or legal person who ~~is involved in the~~decides on the practical deployment of artificial intelligence, robotics and related technologies, ~~and has an~~exercises control opver~~ating or managing function~~ the risk and who benefits from its deployment;

Amendment 511 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point j

(j) ‘deployer’ means ~~any natural or legal person who is involved in the deployment of artificial intelligence, robotics and related technologies, and has an operating or managing func~~the person who decides on the use of the AI-system, exercises control over the associated risk and benefits from its operation;

Amendment 512 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point k

~~(k) ‘use’ means any action relating to artificial intelligence, robotics and related technologies other than development or deployment;~~deleted

Amendment 513 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point l

~~(l) ‘user’ means any natural or legal person who uses artificial intelligence, robotics and related technologies other than for the purposes of development or deployment;~~deleted

Amendment 514 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point m

~~(m) ‘bias’ means any prejudiced or partial personal or social perception of a person or group of persons on the basis of their personal traits;~~deleted

Amendment 515 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point m

(m) ‘bias’ means any prejudiced or ~~partial~~ personal or social perception of a person or group of persons on the basis of their personal traits;

Amendment 516 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point n

~~(n) ‘discrimination’ means any differential treatment of a person or group of persons based on a ground which has no objective or reasonable justification and is therefore prohibited by Union law;~~deleted

Amendment 517 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point o

(o) ‘injury or harm’ means physical, emotional or mental injury, bias, discrimination or stigmatization, suffering caused by a lack of inclusivity and diversity, financial or economic loss, loss of employment or educational opportunity, undue restriction of freedom of choice, wrongful conviction, environmental harm and any infringement of Union law that is detrimental to a person;deleted

Amendment 518 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point o

(o) ‘injury or harm’ means physical~~, emotional~~ or mental injury, ~~bias, discrimination or stigmatization, suffering caused by a lack of inclusivity and diversity~~such as hate speech, loss of privacy bias, financial or economic loss, loss of employment or educational opportunity, undue restriction of freedom of choice, ~~wrongful conviction, environmental harm~~and expression and any infringement of Union law that is detrimental to a person;

Amendment 519 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point o

(o) ‘injury or harm’ means physical, ~~emotion~~psychological or mental injury, bias, discrimination or stigmatization, ~~suffering caused by a lack~~resulting in particular from a violation of inclusivity and diversity, financial or economic loss, loss of employment or educational opportunity, undue restriction of freedom of choice, wrongful conviction, environmental harm and any infringement of Union law that is detrimental to a person;

Amendment 52 #

Motion for a resolution
Recital K

K. whereas each Member State should des~~tablish~~ignate a national supervisory authority responsible for ~~ensuring, assessing and monitoring compliance~~assessing the compliance of high-risk technologies with the ethical framework, and for enabling discussion and exchange of points of view in close cooperation with the concerned stakeholders and the civil society;

Amendment 520 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point p

(p) ‘governance’ means the manner of ensuring that the highest standards and the appropriate protocols of behaviour are adopted and observed by developers, deployers and users, based on a formal set of rules, procedures and values, and which allows them to deal appropriately with ethical matters as or before they arise.deleted

Amendment 521 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point p

Amendment 522 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point p

(p) ‘governance’ means the manner of ensuring that the ~~highest~~appropriate standards and ~~the appropriate~~ protocols of behaviour are adopted and observed by developers, deployers and users, based on a formal set of rules, procedures and values, and which allows them to deal appropriately with ethical matters as or before they arise.

Amendment 523 #

Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point p

Amendment 524 #

Motion for a resolution
Annex I – part B – Article 5 – paragraph 1

Amendment 525 #

Motion for a resolution
Annex I – part B – Article 5 – paragraph 1

Amendment 526 #

Motion for a resolution
Annex I – part B – Article 5 – paragraph 2

2. The development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out in ~~a manner that~~the best interest of citizens and contribute to protect the social, economic and well-being of society by ensuresing that human dignity and the fundamental rights set out in the Charter are fully respected.

Amendment 527 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 5 – paragraph 3

3. The development, deployment and use of development, deployment and use of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out in the best interest of citizens. In particular, the potential of such technologies and the opportunities that they provide shall be taken into consideration having regard at all times to the need to protect and foster the social, environmental and economic well-being of society.deleted

Amendment 528 #

Motion for a resolution
Annex I – part B – Article 5 – paragraph 3

Amendment 529 #

Motion for a resolution
Annex I – part B – Article 5 – paragraph 3

3. The development, deployment and use of development, deployment and use of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out in the best interest of citizens. In particular, the potential of such technologies and the opportunities that they provide shall be taken into consideration, having regard at all times to the centrality and irreplaceability of human beings and the need to protect and foster the social, environmental and economic well-being of society.

Amendment 53 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital K

K. wWhereas ~~each~~the Member State ~~should~~s may, depending on their capacities, consider establishing a national supervisory authority responsible for ensuring, assessing and monitoring compliance, and for enabling discussion and exchange of points of view in close cooperation with the concerned stakeholders and the civil society;

Amendment 530 #

Motion for a resolution
Annex I – part B – after article 5 – chapter title (new)

Chapter II: Obligations for high-risk technologies

Amendment 531 #

Motion for a resolution
Annex I – part B – Article 6 – title

Human-centric ~~and human-made~~ artificial intelligence

Amendment 532 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 6 – paragraph 1

1. Any high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed and used in a human- centric manner with the aim of contributing to the existence of a democratic, pluralistic and equitable society by safeguarding human autonomy and decision-making and ensuring human agency.

Amendment 533 #

Motion for a resolution
Annex I – part B – Article 6 – paragraph 1

1. ~~Any artificial intelligence, robotics and related~~High-risk technologies, including software, algorithms and data used or produced by such technologies, shall: (a) be developed, deployed and used in a human- centric manner with the aim of contributing to the existence of a democratic, pluralistic and equitable society by safeguarding human autonomy and decision-making and ensuring human agency.;

Amendment 534 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 6 – paragraph 2

~~2. The technologies listed in paragraph 1 shall~~(b) be developed, deployed and used in a manner that guarantees full human oversight at any time~~, in particular where that development, deployment or use entails a risk of breaching the ethical principles set out in this Regulation.~~; and

Amendment 535 #

Motion for a resolution
Annex I – part B – Article 6 – paragraph 2

2. The technologies listed in paragraph 1 shall be developed, deployed and used in a manner that guarantees full human oversight at any time~~, in particular where that development, deployment or use entails a risk of breaching the ethical principles set out in this Regulation~~.

Amendment 536 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 6 – paragraph 2

2. The technologies listed in paragraph 1 shall be developed, deployed and used in a manner that guarantees full human oversight at any time, in particular where that development, deployment or use entails a risk of breaching Union law or the ethical principles set out in this Regulation.

Amendment 537 #

Motion for a resolution
Annex I – part B – Article 6 – paragraph 3

~~3. The technologies listed in paragraph 1 shall~~(c) be developed, deployed and used in a manner that allows human control to be regained at any time, including through the altering or halting of those technologies~~, when that development, deployment or use entails a risk of breaching the ethical principles set out in this Regulation~~.

Amendment 538 #

Motion for a resolution
Annex I – part B – Article 6 – paragraph 3

3. The technologies listed in paragraph 1 shall be developed, deployed and used in a manner that allows human control to be regained ~~at any time~~when needed, including through the ~~altering or~~ halting of those technologies~~, when that development, deployment or use entails a risk of breaching the ethical principles set out in this Regulation~~.

Amendment 539 #

Motion for a resolution
Annex I – part B – Article 6 – paragraph 3

3. The technologies listed in paragraph 1 shall be developed, deployed and used in a manner that allows human control to be regained at any time, including through the altering or halting of those technologies, when that development, deployment or use entails a risk of breaching Union law or the ethical principles set out in this Regulation.

Amendment 54 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital K

K. whereas each Member State should des~~tablish a national~~ignate a competent supervisory authority responsible for ensuring, assessing and monitoring compliance, and for enabling discussion and exchange of points of view in close cooperation with the concerned stakeholders and the civil society;

Amendment 540 #

Motion for a resolution
Annex I – part B – Article 7

Annex I – part B – Article 7 1. Regulation, artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, which entail a significant risk of breaching the ethical principles set out in this Regulation shall be considered high-risk technologies. 2. Where artificial intelligence, robotics and related technologies are considered high-risk technologies, an assessment of compliance of those technologies with the obligations set out in this Regulation shall be carried out and monitored by the national supervisory authorities referred to in Annex I – part B – Artdeleted Risk assessment For the purposes of this Without prejudicle 14. 3. the risk assessment of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out on the basis of objective criteria harmonised at Union level and in accordance with applicable sectorial legislation.to paragraph 1,

Amendment 541 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 1

1. For the purposes of this Regulation, artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, ~~which entail a significant risk of breaching the ethical principles set out in this Regulation shall be considered high-risk technologies~~shall be considered high-risk technologies when, considering their intended use and the critical sectors where they are employed in accordance with the Annex to this Regulation, their autonomous operation involves a significant potential to cause harm to one or more persons, in a manner that is random and impossible to predict in advance. Determining the significance of the potential shall depends on the interplay between the severity of possible harm or damage, the likelihood that the risk materializes and the manner in which the AI-system is being used.

Amendment 542 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 1

Amendment 543 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 1 a (new)

1a. The risk assessment of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out by the national supervisory authorities referred to in Article 14 on the basis of a common and exhaustive list of high-risk artificial intelligence, robotics and related technologies in accordance with the objective criteria provided for in paragraph 1 of this Article and elaborated jointly by the national supervisory authorities, the European Commission and other relevant institutions, bodies, offices and agencies of the Union in the context of their cooperation.

Amendment 544 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 2

2. Where artificial intelligence, robotics and related technologies are considered high-risk technologies, an assessment of compliance of those technologies with the obligations set out in this Regulation shall be carried out and monitored by the ~~national~~competent supervisory authorities referred to in Article 14 or the European Agency for Artificial Intelligence.

Amendment 545 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 2 a (new)

2a. Upon request by any developer, deployer or user of high-risk artificial intelligence, robotics and related technologies are considered high-risk technologies seeking to certify the positive assessment of compliance carried out, the respective national supervisory authority shall issue a European certificate of ethical compliance. Such a European certificate of ethical compliance shall be issued in accordance with the common criteria and an application process developed jointly by the national supervisory authorities, the European Commission and other relevant institutions, bodies, offices and agencies of the Union in the context of their cooperation.

Amendment 546 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 3

3. Without prejudice to paragraph 1, the risk assessment of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out on the basis of objective criteria harmonised at Union level and in accordance with applicable sectorial legislation.deleted

Amendment 547 #

Motion for a resolution
Annex I – part B – Article 7 – paragraph 3

3. Without prejudice to paragraph 1, the risk assessment of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be carried out on the basis of a risk-based approach and objective criteria harmonised at Union level and in accordance with applicable sectorial legislation.

Amendment 548 #

Motion for a resolution
Annex I – part B – Article 7 a (new)

Article 7a Voluntary labelling scheme for non high- risk AI technologies 1. For artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies that do not qualify as high-risk and that are not subject to the mandatory requirements and risk assessment established by this Regulation, a voluntary labelling scheme should be established. 2. Under such a voluntary labelling scheme, interested economic operators can decide to make themselves subject either to the requirements listed in this Regulation or to a specific set of similar requirements especially established for the purposes of the voluntary scheme by national authorities. 3. The economic operators concerned shall be awarded a quality label for their artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, provided that those technologies comply with the applicable requirements in accordance with paragraph 2 of this Article.

Amendment 549 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – introductory part

1. Any high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union shall ~~be developed, deployed and used in a manner that ensures they do not breach the ethical principles set out in this Regulation. In particular, they shall~~ be:

Amendment 55 #

Motion for a resolution
Recital K

K. whereas each Member State should establish a national supervisory ~~authority~~system responsible for ensuring, assessing and monitoring compliance, and for enabling discussion and exchange of points of view in close cooperation with the concerned stakeholders and the civil society;

Amendment 550 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point a

~~(a) developed, deployed and used in a consistent manner so that they do not pursue aims or do not carry out activities other than those for which they have been conceived;~~deleted

Amendment 551 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point a

~~(a) developed, deployed and used in a consistent manner so that they do not pursue aims or do not carry out activities other than those for which they have been conceived;~~deleted

Amendment 552 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point b

(b) developed, deployed and used in a resilient manner so that they ensure an adequate level of security by adhering to minimum cybersecurity baselines proportionate to identified risk, and one that prevents any technical vulnerabilities from being exploited for unfair or unlawful purposes;

Amendment 553 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point b

(b) developed, deployed and used in a resilient manner so that they ensure an adequate level of security, and one that prevents any technical vulnerabilities from being exploited for ~~unfair~~malicious or unlawful purposes;

Amendment 554 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point b

Amendment 555 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point c

(c) developed, deployed and used in a secure manner that ensures there are safeguards that include a fall-back plan and action in case of a risk of a ~~breach of the ethical principles set out in this Regulation~~safety or security risk;

Amendment 556 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point d

(d) developed, deployed and used in a manner that ensures that there is trust that the technology works as expected by its user and performance is reliable as regards reaching the aims and carrying out the activities they have been conceived for, including by ensuring that all operations are reproducible;

Amendment 557 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point d

(d) developed, deployed and used in a manner that ensures ~~that there is trust that the performance is reliabl~~are reliable performance as regards reaching the aims and carrying out the activities they have been conceived for, including by ensuring that all operations are reproducible;

Amendment 558 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point d

(d) developed, deployed and used in a manner that ensures that ~~there is trust that~~ the performance is reliable as regards reaching the aims and carrying out the activities they have been conceived for, including by ensuring that all operations are reproducible;

Amendment 559 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point e

(e) developed, deployed and used in a manner that ensures that the performance of the aims and activities of the particular technologies is accurate; if occasional inaccuracies cannot be avoided, the system shall indicate, to the extent possible, the likeliness of errors and inaccuracies to deployers and users through an appropriate disclaimer message;

Amendment 56 #

Motion for a resolution
Recital K

K. whereas each Member State should establish a national supervisory authority responsible for ensuring, assessing and monitoring compliance, and for enabling discussion and exchange of points of view in close cooperation with the concerned stakeholders and the civil society and national supervisory authorities should cooperate with each other;

Amendment 560 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point e

(e) developed, deployed and used in a manner that ensures that the performance of the aims and activities of the particular technologies is accurate; if occasional inaccuracies cannot be avoided, the system shall indicate the likeliness of errors and inaccuracies to deployers and users ~~through an appropriate disclaimer message~~appropriately;

Amendment 561 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point f

(f) developed, deployed and used in an easily explainable, auditable, traceable, and transparent manner so as to ensure that there can be a review of the technical processes of the technologies;

Amendment 562 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point g

(g) developed, deployed and used in a manner such that they ~~are capable of warning~~inform users that they are interacting with artificial intelligence systems, duly and comprehensively disclosing their capabilities, accuracy and limitations to ~~artificial intelligence~~ developers, deployers and users of high- risk technologies;

Amendment 563 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point g

(g) developed, deployed and used in a manner such that they are capable of warning users that they are interacting with artificial intelligence systems, duly disclosing their capabilities, accuracy and limitations to ~~artificial intelligence developers,~~ deployers and users;

Amendment 564 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point h

(h) in accordance with Article 6(3), developed, deployed and used in a manner that makes it possible, in the event of non- compliance with the safety features set out in subparagraphs (a) to (g), for the technologies concerned to be temporarily disabled and to revert to ~~historical~~a previous state, identified as restoring safe functionalities.

Amendment 565 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 2

2. In accordance with Article 6(2b), ~~the~~high-risk technologies ~~mentioned in paragraph 1~~, including software, algorithms and data used or produced by such technologies shall be developed, deployed and used in transparent and traceable manner so that their elements, processes and phases are documented to the highest ~~standards, and that it is possible for the national supervisory authorities referred to in Article 14 to assess the compliance of such technologies with the obligations set out in this Regulation~~applicable standards. In particular, the developer, deployer or user of those technologies shall be responsible for, and be able to demonstrate, compliance with the safety features set out in paragraph 1.

Amendment 566 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 2

2. In accordance with Article 6(2), the technologies mentioned in paragraph 1 shall be developed, deployed and used in transparent and traceable manner so that their elements, processes and phases are documented to the highest possible standards, and that it is possible for the national supervisory authorities referred to in Article 14 to assess the compliance of such technologies with the obligations set out in this Regulation. In particular, the developer, deployer or user of those technologies shall be responsible for, and be able to demonstrate, compliance with the safety features set out in paragraph 1.

Amendment 567 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 2

2. In accordance with Article 6(2), the technologies mentioned in paragraph 1 shall be developed, deployed and used in transparent and traceable manner so that their elements, processes and phases are documented to the highest standards, and that it is possible for the ~~national~~competent supervisory authorities referred to in Article 14 to assess the compliance of such technologies with the obligations set out in this Regulation. In particular, the developer, deployer or user of those technologies shall be responsible for, and be able to demonstrate, compliance with the safety features set out in paragraph 1.

Amendment 568 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 3

3. The developer, deployer or user of the technologies mentioned in paragraph 1 shall ensure that the measures taken to ensure compliance with the ~~safety~~ features set out in paragraph 1 can be audited by the ~~national~~competent supervisory authorities referred to in Article 14 and other supervisory bodies.

Amendment 569 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 3

3. The developer, deployer or user of ~~the~~high-risk technologies ~~mentioned in paragraph 1~~ shall ensure that the measures taken to ensure compliance with the safety features set out in paragraph 1 can be audited by the national supervisory authorities referred to in Article 14.

Amendment 57 #

Motion for a resolution
Recital L

L. whereas Parliament continues to call for the establishment of a European Agency to ensure a harmonised approach across the Union and address the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments.deleted

Amendment 570 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 8 – paragraph 4

4. Users shall be presumed to have complied with the obligations set out in this Article where their use of artificial, robotics and related technologies, including software, algorithms and data used or produced by such technologies, is carried out in good faith and in no way contravenes the ethical principles laid down in this Regulation.deleted

Amendment 571 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 4

Amendment 572 #

Motion for a resolution
Annex I – part B – Article 8 – paragraph 4

4. Users shall be presumed to have complied with the obligations set out in this Article where their use of artificial, robotics and related technologies, including software, algorithms and data used or produced by such technologies, is carried out in good faith in accordance with the safety and use instructions provided by the developed and/or the deployer and in no way contravenesing the ethical principles laid down in this Regulation.

Amendment 573 #

Motion for a resolution
Annex I – part B – Article 9 – paragraph 1

1. Any high-risk software, algorithm or data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be such as to ensure respect for human dignity and equal treatment for all in line with Union law.

Amendment 574 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 9 – paragraph 1

1. Any software, algorithm or data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be such as to ensure respect for human rights and dignity and equal treatment for all.

Amendment 575 #

Motion for a resolution
Annex I – part B – Article 9 – paragraph 1

1. ~~Any~~High-risk technologies, including software, algorithm or data used or produced by ~~artificial intelligence, robotics and related~~such technologies developed, deployed or used in the Union shall be such as to ensure respect for human dignity and equal treatment for all.

Amendment 576 #

Motion for a resolution
Annex I – part B – Article 9 – paragraph 2

2. Any software, algorithm or data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be unbiased and, without prejudice to paragraph 3, shall not discriminate on grounds such as race, gender, sexual orientation, pregnancy, disability, physical or genetic features, age, national minority, ethnic or social origin, language, religion or belief, political views or civic participation, citizenship, civil or economic status, education, or criminal record, and shall not seek to promote certain ideas, values and uniform views of society.

Amendment 577 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 9 – paragraph 2

2. Any high-risk software, algorithm or data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be unbiased and, without prejudice to paragraph 3, shall not discriminate on grounds such as race, gender, sexual orientation, pregnancy, disability, physical or genetic features, age, national minority, ethnic or social origin, language, religion or belief, political views or civic participation, citizenship, civil or economic status, education, or criminal record.

Amendment 578 #

Motion for a resolution
Annex I – part B – Article 9 – paragraph 2

2. ~~Any~~High-risk technologies, including software, algorithm ors and data used or produced by ~~artificial intelligence, robotics and related~~such technologies, developed, deployed or used in the Union shall be unbiased and, without prejudice to paragraph 3, shall not discriminate on grounds such as race, gender, sexual orientation, pregnancy, disability, physical or genetic features, age, national minority, ethnic or social origin, language, religion or belief, political views or civic participation, citizenship, civil or economic status, education, or criminal record.

Amendment 579 #

Gilles LEBRETON, Jean-Paul GARRAUD, Gunnar BECK, Luisa REGIMENTI

Motion for a resolution
Annex I – part B – Article 9 – paragraph 2

2. Any software, algorithm or data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be unbiased and~~, without prejudice to paragraph 3,~~ shall not discriminate on grounds such as ~~race~~perceived ethnicity, gender, sexual orientation, pregnancy, disability, physical or genetic features, age, national minority, ethnic or social origin, language, religion or belief, political views or civic participation, citizenship, civil or economic status, education, or criminal record.

Amendment 58 #

Motion for a resolution
Recital L

Amendment 580 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 1

1. ~~Any artificial intelligence, robotics and related~~High-risk technologies, ,including software, algorithms and data used or produced by such technologies, shall be developed, deployed and used in the Union in compliance with the relevant Union law, principles and values~~, in a manner that ensures optimal social, environmental and economic outcomes and that does not result in injury or harm of any kind to being caused to individuals or society~~..

Amendment 581 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 1

1. Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed and used in the Union in compliance with the relevant Union law, principles and values, in a manner that ensures optimal social, environmental and economic outcomes and that does not result in injury or harm of any kind to being caused to individuals or society, paying particular attention to the possible adverse effects on employment and workers’ rights.

Amendment 582 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 1

1. Any high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed and used in the Union in ~~compliance with the relevant Union law, principles and values, in a~~a socially responsible manner that ensures ~~optimal social~~social well-being, environmental and economic outcomes that contributes to gender balance and that does not result in injury or harm of any kind to being caused to individuals or society in compliance with relevant Union laws, principles and values.

Amendment 583 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2

2. Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union shall be developed, deployed and used in a socially responsible manner. In particular, such a manner shall mean that such technologies are: (a) manner that contributes to improving individual development, collective well- being and the healthy functioning of democracy, without interfering in political processes, decision-making and elections or contributing to the dissemination of disinformation; (b) manner that contributes to the achievement of a fair society by helping to increase citizens’ health and well-being, fostering equality in the creation and availability of economic, social and political opportunity and respecting workers’ rights; (c) developed, deployed and used in a manner that contributes to public debate, complements and empowers human cognitive skills, encourages quality education and promotes multilingualism while reflecting the cultural diversity of the Union; (d) developed, deployed and used in a gender-balanced manner that narrows the gender gap by providing equal opportunities for all; (e) manner that contributes to the narrowing of the digital divide among regions, age groups and social classes, the promotion of digital literacy and skills, innovation and creativity, while respecting intellectual property rights;deleted developed, deployed and used in a developed, deployed and used in a developed, deployed and used in a

Amendment 584 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – introductory part

2. ~~Any artificial intelligence, robotics and related~~High-risk technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union shall be developed, deployed and used in a socially responsible manner. ~~In particular, such a manner shall mean that such technologies are:~~

Amendment 585 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point a

(a) developed, deployed and used in a manner that contributes to improving individual development, collective well- being and the healthy functioning of democracy, without interfering in political processes, decision-making and elections or contributing to the dissemination of disinformation;deleted

Amendment 586 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point a

Amendment 587 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point b

(b) developed, deployed and used in a manner that contributes to the achievement of a fair society by helping to increase citizens’ health and well-being, fostering equality in the creation and availability of economic, social and political opportunity and respecting workers’ rights;deleted

Amendment 588 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point b

Amendment 589 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point c

(c) developed, deployed and used in a manner that contributes to public debate, complements and empowers human cognitive skills, encourages quality education and promotes multilingualism while reflecting the cultural diversity of the Union;deleted

Amendment 59 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital L

Amendment 590 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point c

Amendment 591 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point c

Amendment 592 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point d

~~(d) developed, deployed and used in a gender-balanced manner that narrows the gender gap by providing equal opportunities for all;~~deleted

Amendment 593 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point d

~~(d) developed, deployed and used in a gender-balanced manner that narrows the gender gap by providing equal opportunities for all;~~deleted

Amendment 594 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e

Amendment 595 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e

Amendment 596 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e

(e) developed, deployed and used in a manner that contributes to the narrowing of the digital divide among regions, age groups and social classes, the promotion of digital literacy and skills, innovation and creativity, while respecting intellectual property rights and exploring the potential of these technologies, namely through their explainability, in detecting infringements of such rights;

Amendment 597 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e

Amendment 598 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA, Liesje SCHREINEMACHER

Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e

(e) developed, deployed and used in a manner that contributes to the narrowing of the digital divide among regions, age groups and social classes, the protection of children, the promotion of digital literacy and skills, innovation and creativity, while respecting intellectual property rights;

Amendment 599 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 3

3. The Union and its Member States shall encourage research projects intended to provide solutions, based on artificial intelligence, robotics and related technologies, that seek to promote social inclusion, plurality, solidarity, fairness, equality and cooperation.deleted

Amendment 6 #

Motion for a resolution
Citation 14 a (new)

- having regard to the OECD Council Recommendation on Artificial Intelligence adopted on 22 May 2019,

Amendment 60 #

Motion for a resolution
Recital L

Amendment 600 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 3

3. The Union and its Member States shall encourage research projects intended to provide solutions, based on artificial intelligence, robotics and related technologies, that seek to promote social inclusion, democracy, plurality, solidarity, fairness, equality and cooperation.

Amendment 601 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 10 – paragraph 4

4. The social effects of the ubiquitous presence of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union shall be monitored by the national supervisory authorities referred to in Article 14, in order to avoid disruptive effects on social agency and social relationships, as well as the deterioration of social skills.deleted

Amendment 602 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 4

Amendment 603 #

Motion for a resolution
Annex I – part B – Article 10 – paragraph 4

4. The social effects of the ubiquitous presence of artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed or used in the Union shall be monitored by the ~~national~~competent supervisory authorities referred to in Article 14 and other supervisory bodies, in order to avoid disruptive effects on social agency and social relationships, as well as the deterioration of social skills.

Amendment 604 #

Motion for a resolution
Annex I – part B – Article 11 – title

Environmental ~~friendliness and~~ sustainability

Amendment 605 #

Motion for a resolution
Annex I – part B – Article 11 – paragraph 1

1. Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed or used in the Union in compliance with Union law, principles and values, in a manner that ~~ensures optimal environmentally friendly outcomes and minimis~~preserves the environment, minimises and remedies their environmental footprint during their lifecycle and through their entire supply chain, in order to contribute to the green transition, support the achievement of climate neutrality and circular economy goals.

Amendment 606 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 11 – paragraph 1

1. Any high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed or used in the Union ~~in compliance with Union law, principles and values,~~ in a manner that ensures ~~optimal~~ environmentally ~~friendly~~sustainable outcomes and minimises their environmental footprint during their lifecycle and through their entire supply chain, in order to support the achievement of climate neutrality and circular economy goals in accordance with the applicable Union law.

Amendment 607 #

Motion for a resolution
Annex I – part B – Article 11 – paragraph 1

1. ~~Any artificial intelligence, robotics and related~~High-risk technologies, including software, algorithms and data used or produced by such technologies, shall be developed, deployed or used in the Union in compliance with Union law~~, principles and values, in a manner that ensures optimal environmentally friendly outcomes and~~ and commitments related to the protection of the environment and where applicable they shall pursue the objective of minimisesing their environmental footprint during their lifecycle and through their entire supply chain, in order to support the achievement of climate neutrality and circular economy goals.

Amendment 608 #

Stéphane SÉJOURNÉ, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 11 – paragraph 2

2. The Union and its Member States shall encourage and promote research projects intended to provide solutions, based on artificial intelligence, robotics and related technologies, seeking to address environmental issues such as waste production, the carbon footprint, climate change and environmental degradation.deleted

Amendment 609 #

Motion for a resolution
Annex I – part B – Article 11 – paragraph 2

2. The Union and its Member States shall encourage and promote research projects intended to provide solutions, based on artificial intelligence, robotics and related technologies, seeking to ~~address~~remedy environmental issues such as waste production, the carbon footprint, climate change and environmental degradation.

Amendment 61 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital L

L. whereas ~~Parliament continues to call for the establishment of a European Agency to ensure a harmonised approach across the Union and~~in order to ensure a harmonised approach across the Union and the optimal functioning of the Digital Single Market and to address the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments, Parliament calls for the establishment of a European agency for artificial intelligence, robotics and related technologies.

Amendment 610 #

Motion for a resolution
Annex I – part B – Article 11 – paragraph 3

3. Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be assessed as to their environmental friendliness and sustainability by the national supervisory authorities, referred to in Article 14, ensuring that measures are put in place to mitigate their general impact as regards natural resources, energy consumption, waste production, the carbon footprint, climate change and environmental degradation.deleted

Amendment 611 #

Motion for a resolution
Annex I – part B – Article 11 – paragraph 3

3. Any high-risk artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be assessed as to their environmental ~~friendliness and~~ sustainability by the national supervisory authorities, referred to in Article 14, ensuring that measures are put in place to mitigate their general impact as regards natural resources, energy consumption, waste production, the carbon footprint, climate change and environmental degradation in order to ensure compliance with the applicable Union or national law.

Amendment 612 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 11 – paragraph 3

3. Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, shall be assessed as to their environmental friendliness and sustainability by the ~~national~~competent supervisory authorities, referred to in Article 14 and other supervisory bodies, ensuring that measures are put in place to mitigate their general impact as regards natural resources, energy consumption, waste production, the carbon footprint, climate ~~change~~emergency and environmental degradation.

Amendment 613 #

Motion for a resolution
Annex I – part B – Article 11 a (new)

Article 11a Compliance assessment 1. The national supervisory authorities referred to in Article 14 shall assess whether high-risk technologies are compliant with the obligations set out in Chapter II of this Regulation. 2. The software, algorithms and data used or produced by high-risk technologies which have been assessed compliant within the meaning of this article shall be presumed compliant with this Regulation, unless the national supervisory authority decides to conduct an assessment at its own initiative or at the request of the developer, the deployer or the user. 3. When assessing the compliance with the obligations specified under Articles 9, 10 and 11, the national authority shall in particular consider the purposes pursued by the development, the deployment or the use the technologies, especially when it relates to: (a) individual development, collective well-being, democracy and political processes; (b) health and workers’ rights; (c) education. 4. The Commission shall prepare binding guidelines, in accordance with the applicable sectorial legislation, on the methodology to be used by the national supervisory authorities for the compliance assessment by the date of the entry into force of this Regulation.

Amendment 614 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 11 b (new)

Article 11b Implementation guidance The Commission shall prepare non- binding guidelines on the methodology for compliance with this Regulation intended to developers, deployers and users. In doing so, the Commission shall consult relevant stakeholders. The Commission shall publish the guidelines by the date of the entry into force of this Regulation.

Amendment 615 #

Sergey LAGODINSKY, Alexandra GEESE, Patrick BREYER

Motion for a resolution
Annex I – part B – Article 12 – title

Privacy ~~and biometric recognition~~

Amendment 616 #

Motion for a resolution
Annex I – part B – Article 12 – paragraph 1 a (new)

1a. The use and gathering of biometric data for remote identification purposes for deployment of facial recognition in public area, carries specific risk for fundamental rights and shall be limited only to substantial public interest in accordance with EU data protection rules and, in particular, the GDPR. In that case, the processing must take place on the basis of Union or national law, subject to the requirements of proportionality, respect for the essence of the right to data protection and appropriate safeguards and in compliance with Article 12(1).

Amendment 617 #

Motion for a resolution
Annex I – part B – Article 12 – paragraph 2

2. In accordance with Article 5(2), where remote recognition technologies, such as biometric recognition, are deployed or used by Member States’ public authorities for the purpose of responding to a national emergency, those authorities shall ensure that such deployment or use is limited to specific objectives, restricted in time and carried out with due regard for human dignity and the fundamental rights set out in the Charter.deleted

Amendment 618 #

Motion for a resolution
Annex I – part B – Article 12 – paragraph 2

2. In accordance with Article 5(2), where remote recognition technologies, such as biometric recognition, are deployed or used by Member States’ public authorities ~~for the purpose of responding to a national emergency, those authorities~~, they shall ensure that such deployment or use is voluntary, proportionate, targeted and limited to specific objectives, restricted in time and carried out ~~with due regard for human dignity and the fundamental rights set out in the Charter~~in accordance with Union law, with due regard for human rights and dignity, autonomy and self-determination of the individual and the fundamental rights set out in the Charter. Criteria for and limits to their use should be subjected to judicial review and submitted to democratic scrutiny and debate involving civil society, in order to determine whether, when and how it should take place, and such use should be mindful of its psychological and sociocultural impact.

Amendment 619 #

Motion for a resolution
Annex I – part B – Article 12 – point 2

2. In accordance with Article 5(2), where remote recognition technologies, such as biometric recognition, are deployed or used by Member States’ public authorities for the purpose of responding to a national emergency, those authorities shall ensure that such deployment or use is limited to specific objectives and locations (some establishments might be visited by the public and pose, by their nature, a high security risk to individuals and to the State), restricted in time and carried out with due regard for human dignity and the fundamental rights set out in the Charter.

Amendment 62 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA, Liesje SCHREINEMACHER

Motion for a resolution
Recital L

L. whereas ~~Parliament continues to call for the establishment of a European Agency to~~the idea of entrusting an existing or new European Agency or European body with the task of ensureing a harmonised approach across the Union ~~and addres~~should be assessed as regards the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments.

Amendment 620 #

Motion for a resolution
Annex I – part B –After Article 12 – chapter title (new)

Chapter III: Specific requirements

Amendment 621 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER

Motion for a resolution
Article 12 a (new)

Article 12a Public power decisions 1. Member States shall conduct an impact assessment on fundamental rights for high-risk technologies used within their prerogatives of public powers that have a significant and direct impact on the rights and obligations of natural or legal persons. 2. High-risk technologies that have a direct and significant impact on rights and obligations of natural and legal persons shall be subject to strict human verification, and due process.

Amendment 622 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA, Liesje SCHREINEMACHER

Motion for a resolution
Annex I – part B – Article 12 b (new)

Article 12b Right to redress 1. Any natural or legal person shall be able to seek redress for damages caused by a decision issued at her/his detriment by high-risk technologies.

Amendment 623 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Karen MELCHIOR

Motion for a resolution
Annex I – part B –After Article 12 b – chapter title (new)

Chapter IV: Institutional oversight

Amendment 624 #

Motion for a resolution
Annex I – part B – Article 13 – paragraph 1

1. Artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall comply with relevant governance standards established by the national supervisory authorities referred to in Article 14 in accordance with Union law, principles and values.deleted

Amendment 625 #

Motion for a resolution
Annex I – part B – Article 13 – paragraph 1

1. Artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall comply with relevant governance standards ~~establish~~drafted by the European Commission or another relevant institution, body, office or agency of the Union designated for this task and implemented by the national supervisory authorities referred to in Article 14 in accordance with Union law, principles and values.

Amendment 626 #

Motion for a resolution
Annex I – part B – Article 13 – paragraph 1

1. Artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall comply with relevant governance standards established by the ~~national~~competent supervisory authorities referred to in Article 14 and further supervisory bodies, in accordance with Union law, principles and values.

Amendment 627 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 13 – paragraph 2

2. Data ~~used or produc~~produced or destined to be used by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall be managed by developers, deployers and users in accordance with relevant standards referred to in paragraph 1, as well as with relevant industry and business protocols. In particular, developers and deployers shall carry out, where feasible, quality checks of the external sources of data used by artificial intelligence, robotics and related technologies, and shall put oversight mechanisms in place regarding their collection, storage, processing and use.

Amendment 628 #

Motion for a resolution
Annex I – part B – Article 13 – paragraph 2

2. Data used or produced by ~~artificial intelligence, robotics and related~~high-risk technologies developed, deployed or used in the Union shall be managed by developers, deployers and users in accordance with the relevant ~~standards referred to in paragraph 1~~national, European and international rules, as well as with relevant industry and business protocols. In particular, developers and deployers shall carry out, where feasible, quality checks of the external sources of data used by artificial intelligence, robotics and related technologies, and shall put oversight mechanisms in place regarding their collection, storage, processing and use.

Amendment 629 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 13 – paragraph 3

3. Without prejudice to portability rights and rights of persons whose usage of ~~artificial intelligence, robotics and related~~high-risk technologies has generated data, the collection, storage, processing, sharing of and access to data used or produced by artificial intelligence, robotics and related technologies developed, deployed or used in the Union shall comply with the relevant ~~standards referred to in paragraph 1~~national, European and international rules, as well as with relevant industry and business protocols. In particular, developers and deployers shall ensure those protocols are applied during the development and deployment of ~~artificial intelligence, robotics and related~~high-risk technologies, by clearly defining the requirements for processing and granting access to data used or produced by these technologies, as well as the purpose, scope and addressees of the processing and the granting of access to such data, all of which shall at all times be auditable and traceable.

Amendment 63 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital L

L. whereas ~~Parliament continues to call for the establishment of a European Agency to~~the idea of entrusting an existing European body with the task of ensureing a harmonised approach across the Union ~~and addres~~should be assessed as regards the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments.

Amendment 630 #

Motion for a resolution
Annex I – part B – Article 13 a (new)

Article 13a Guidelines The Commission is drawing up guidelines for the application of this Regulation, in particular to enable small and medium- sized enterprises to effectively comply with the requirements of Articles 9 to 13 of this Regulation while minimising administrative and other burdens.

Amendment 631 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Adrián VÁZQUEZ LÁZARA, Liesje SCHREINEMACHER

Motion for a resolution
Annex I – part B – Article 14 – paragraph 1

1. Each Member State shall designate an independent public authority to be responsible for monitoring the application of this Regulation (‘supervisory authority’). In accordance with Article 7~~(1) and (2)~~, each national supervisory authority shall be responsible for assessing whether artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed and used in the Union are high- risk technologies and, if so, for assessing and monitoring their compliance with the ethical principles set out in this Regulation.

Amendment 632 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 1

1. Each Member State shall designate an independent public authority to be responsible for monitoring the application of this Regulation (‘supervisory authority’). In accordance with Article ~~7(1) and~~ (2), each national supervisory authority shall be responsible for assessing ~~whether artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed and used in the Union are~~and monitoring the compliance of high- risk technologies ~~and, if so, for assessing and monitoring their compliance~~ with the ethical principles set out in this Regulation.

Amendment 633 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 1

1. Each Member State shall designate an independent public authorityies or authorities with a public purpose to be responsible for monitoring the application of this Regulation ~~(‘supervisory authority’)~~. In accordance with Article 7(1) and (2), each ~~national~~competent supervisory authority shall be responsible for assessing whether artificial intelligence, robotics and related technologies, including software, algorithms and data ~~used or produc~~produced or destined to be used by such technologies, developed, deployed and used in the Union are high-risk technologies and, if so, for assessing and monitoring their compliance with the law and the ethical principles set out in this Regulation.

Amendment 634 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 1

1. Each Member State shall designate an independent public authority to be responsible for monitoring the application of this Regulation (‘supervisory authority’). In accordance with Article 7(1) and (2), each national supervisory authority shall be responsible for assessing whether artificial intelligence, robotics and related technologies, including software, algorithms and data used or produced by such technologies, developed, deployed and used in the Union are high-risk technologies in view of the ethical principles set out in this Regulation and, if so, for assessing and monitoring their compliance with the ~~ethical~~se principles ~~set out in this Regulation~~.

Amendment 635 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 14 – paragraph 1 a (new)

1a. Each national supervisory authority shall supervise the implementation of the legislation in force and, where applicable, the enforceable sectorial regulations arising when a technology is identified as a high-risk technology in view of the ethical principles set out in this Regulation. For this purpose, the supervisory authorities shall cooperate with the sectorial authorities in charge of assessing and monitoring artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies and enforcing their compliance in the light of sectorial legislation.

Amendment 636 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 2

2. Each national supervisory authority shall contribute to the consistent application of this Regulation throughout the Union. For that purpose, the supervisory authorities in each Member State shall cooperate with each other, the Commission and other relevant institutions, bodies, offices and agencies of the Union~~, in particular as regards establishing the governance standards referred to in Article 13(1)~~.

Amendment 637 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 2

2. Each ~~national~~competent supervisory authority 2. shall contribute to the consistent application of this Regulation throughout the Union. For that purpose, the supervisory authorities in each Member State shall cooperate with each other, the Commission and other relevant institutions, bodies, offices and agencies of the Union, in particular as regards establishing the governance standards referred to in Article 13(1).

Amendment 638 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 2 a (new)

2a. The competent supervisory authorities shall serve as a first point of contact in cases of suspected discriminating treatment or of violation of other rights as a result of use of artificial intelligence and automated decision making systems, conduct ethical evaluation of such cases in cooperation with other competent authorities in the Union, notably the Consumer Protection Cooperation Network, national consumer protection bodies and civil society; to facilitate means to individuals to meaningfully contest and remedy harm caused by such discrimination or violations.

Amendment 639 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 3

3. Each national supervisory authority shall be responsible for supervising the application of governance standards to artificial intelligence, robotics and related technologies, including by liaising with the maximum possible number of stakeholders. For that purpose, the supervisory authorities in each Member State shall provide a forum for regular exchange with stakeholders.deleted

Amendment 64 #

Motion for a resolution
Recital L

L. whereas Parliament ~~continues to~~ calls for ~~the establishment of a European Agency to ensure~~ a harmonised approach to be ensured across the Union and ~~address~~for the new opportunities and challenges to be addressed within its framework, in particular those of a cross- border nature, arising from ongoing technological developments.

Amendment 640 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 3

3. Each ~~national~~competent supervisory authority shall be responsible for supervising the application of governance standards to artificial intelligence, robotics and related technologies, including by liaising with the maximum possible number of stakeholders. For that purpose, the supervisory authorities in each Member State shall provide a forum for regular exchange ~~with stakeholders~~and the cooperation and collaboration between stakeholders from academia, research, industry, civil society and individual experts .

Amendment 641 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 14 – paragraph 4

4. Each ~~national~~competent supervisory authority shall provide professional and administrative guidance and support on the general implementation of Union law regarding artificial intelligence, robotics and related technologies, and the ethical principles set out in this Regulation, including to relevant research and development, and small and medium-sized enterprises or start-ups.

Amendment 642 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 4

4. Each national supervisory authority shall provide professional and administrative guidance ~~and support on the general implementation of the ethical principles~~for the implementation of the harmonised ethical framework set out in this Regulation, ~~including~~especially to small and medium-sized enterprises or start-ups.

Amendment 643 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Annex I – part B – Article 14 – paragraph 6

6. ~~Member States shall take all measures necessary to ensure the implementation of the ethical principles set out in this Regulation.~~ Member States shall support relevant stakeholders and civil society, at both Union and national level, in their efforts to ensure a timely, ethical and well- informed response to the new opportunities and challenges, in particular those of a cross-border nature, arising from technological developments relating to artificial intelligence, robotics and related technologies.

Amendment 644 #

Motion for a resolution
Annex I – part B – Article 14 – paragraph 6 a (new)

6a. For the purpose of ensuring the application of ethical principles throughout the territory of the Union as well as a harmonised functioning of the Digital Single Market in the field of artificial intelligence, robotics and related technologies, a European Agency has been established. Together with national supervisory authorities, the Agency will develop common criteria and an application process relating to the granting of a European certificate of ethical compliance following a request by any developer, deployer or user seeking to certify the positive assessment of compliance carried out by the respective national supervisory authority;

Amendment 645 #

Motion for a resolution
Annex I – part B – Article 15

~~Directive (EU) 2019/19373 of the European Parliament and of the Council3 shall apply to the reporting of breaches of this Regulation and the~~Article 15 deleted Reporting of breaches and protection of ~~persons~~ reporting such breaches. __________________ 3Directive (EU) 2019/1937 of the European Parliament and of the Council of 23 October 2019 on the protection of persons who report breaches of Union law (OJ L 305, 26.11.2019, p. 17).persons

Amendment 646 #

Motion for a resolution
Annex I – part B – Article 15 – paragraph 1

Directive (EU) 2019/1937 of the European Parliament and of the Council3 shall apply to the reporting of breaches of this Regulation and the protection of persons reporting such breaches. __________________ 3Directive (EU) 2019/1937 of the European Parliament and of the Council of 23 October 2019 on the protection of persons who report breaches of Union law (OJ L 305, 26.11.2019, p. 17).deleted

Amendment 647 #

Motion for a resolution
Annex I – part B – Article 16

Amendment to Directive (EU) No Directive (EU) No 2019/1937 is amended as follows: (1) In Article 2(1), the following point is added: ‘ ‘(xi) development, deployment and use artificial intelligence, robotics and related technologies.’ (2) In Part I of the Annex, the following point is added: ‘K. Point (a)(xi) of Article 2(1) - development, deployment and use artificial intelligence, robotics and related technologies. “(xxi) Regulation [XXX] of the European Parliament and of the Council on ethical principles for the development, deployment and use artificial intelligence, robotics and related technologies”.’rticle 16 deleted 2019/1937

Amendment 648 #

Motion for a resolution
Annex I – part B – Article 16 – paragraph 1 – point 1

~~(1) In Article 2(1), the following point is added: ‘ ‘(xi) development, deployment and use artificial intelligence, robotics and related technologies.’~~deleted

Amendment 649 #

Motion for a resolution
Annex I – part B – Article 16 – paragraph 1 – point 2

(2) In Part I of the Annex, the following point is added: ‘K. Point (a)(xi) of Article 2(1) - development, deployment and use artificial intelligence, robotics and related technologies. “(xxi) Regulation [XXX] of the European Parliament and of the Council on ethical principles for the development, deployment and use artificial intelligence, robotics and related technologies”.’deleted

Amendment 65 #

Motion for a resolution
Recital L

L. whereas ~~Parliament continues to call for the establishment of a European Agency to~~the European Commission should ensure a harmonised approach across the Union and address the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments.

Amendment 650 #

Motion for a resolution
Annex I – part B – Article 17 – paragraph 1

TIn conjunction with the European Agency for artificial intelligence, robotics and related technologies, the Commission shall keep under regular review the development of artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, and shall by [OJ: please enter the date three years after entry into force], and every three years thereafter, submit to the European Parliament, the Council and the European Economic and Social Committee a report on the application of this Regulation, including an assessment of the possible modification of the scope of application of this Regulation.

Amendment 651 #

Motion for a resolution
Annex I – part B – Article 17 a (new)

Article 17a Exercise of delegation 1. The power to adopt delegated acts is conferred on the Commission subject to the conditions laid down in this Article. 2. The power to adopt delegated acts referred to in Article 2(3) shall be conferred on the Commission for an indeterminate period of time from (date of entry into force of this Regulation). 3. The delegation of power referred to in Article 2(3) may be revoked at any time by the European Parliament or by the Council. A decision to revoke shall put an end to the delegation of the power specified in that decision. It shall take effect the day following the publication of the decision in the Official Journal of the European Union or a later date specified therein. It shall not affect the validity of any delegated act already in force. 4. Before adopting a delegated act, the Commission shall consult experts designated by each Member State in accordance with the principles laid down in the Interinstitutional Agreement of 13 April 2016 on Better Law Making. 5. As soon as it adopts a delegated act, the Commission shall notify it simultaneously to the European Parliament and to the Council. 6. A delegated act adopted pursuant to Article 2(3) shall enter into force only if no objection has been expressed either by the European Parliament or the Council within a period of three months of notification of that act to the European Parliament and the Council or, if, before the expiry of that period, the European Parliament and the Council have both informed the Commission that they will not object. That period shall be extended by three months at the initiative of the European Parliament or of the Council.

Amendment 652 #

Motion for a resolution
Annex I – part B – Annex (new)

ANNEX The two limbed-test to determine if an AI application is ‘high-risk’ is set out as follows: the AI application is employed in a sector where significant risks can be expected to occur given the nature of activities typically undertaken (such as healthcare, transport, energy and parts of the public sector i.e. migration, social security); and the AI application is used in a manner where significant risks are likely to arise (for example, uses of AI that produce legal or other significant effects for the rights of an individual, or that pose a risk to injury death or immaterial damage).

Amendment 66 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Subheading 2

Human-centric ~~and human-made~~ artificial intelligence

Amendment 67 #

Motion for a resolution
Paragraph -1 (new)

-1. Believes that any legislative action, in particular, related to new technologies should be in line with the principles of necessity and proportionality; points out, in this respect, that the ethical framework considered in this report should be applicable to high-risk artificial intelligence, robotics and related technologies;

Amendment 68 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph -1 a (new)

-1a. Considers that such an approach will allow companies to introduce innovative products into the market and create new opportunities while ensuring the protection of the European values;

Amendment 69 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Paragraph -1 b (new)

-1b. Considers that artificial intelligence, robotics and related technologies should be considered« high risk technologies » when they are used in sectors, where given the characteristics of the activities typically undertaken, and are used in such a manner that significant risks can be expected to occur from the viewpoint of safety and fundamental rights and freedoms;

Amendment 7 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA, Javier NART

Motion for a resolution
Recital A

A. whereas so-called ‘artificial intelligence’, or extended intelligence, robotics and related technologies ~~with~~are being developed quickly and can have the potential to directly impact all aspects of our societies positively, as well as negatively, depending on how they are applied, and who can benefit from access to them, including basic social and economic principles and values~~, are being developed very quickly~~;

Amendment 70 #

Motion for a resolution
Paragraph -1 c (new)

-1c. Asks the Commission to establish an exhaustive list of the technologies fulfilling these criteria in the form of annex to the Regulation on ethical principles for the development, deployment and use of artificial intelligence, robotics and related technologies; considers that the Commission should review the exhaustive list, if necessary, every six months by means of a delegated act to add new high- risk technologies or delete existing ones if they do not fulfil the criteria anymore; recalls that any changes to the annex should be thoroughly assessed and justified;

Amendment 71 #

Motion for a resolution
Paragraph 1

1. Declares that the development, deployment and use of artificial intelligence, robotics and related technologies, including but not exclusively by human beings, should always respect human agency and democratic oversight, as well as allow the retrieval of human control at any time;

Amendment 72 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 1

1. Declares that the development, deployment and use of high-risk artificial intelligence, robotics and related technologies, including but not exclusively by human beings, should always respect human agency and oversight, as well as allow the retrieval of human control ~~at any time~~when needed;

Amendment 73 #

Motion for a resolution
Paragraph 1

Amendment 74 #

Motion for a resolution
Paragraph 1

1. Declares that the development, deployment and use of artificial intelligence, robotics and related technologies~~, including but not exclusively by human beings, should always respect~~ should always be ethically guided, and designed to allow for human agency ~~and~~, oversight, ~~as well as allow the retrieval of hum~~understanding and control at any time;

Amendment 75 #

Motion for a resolution
Paragraph 1 a (new)

1a. Considers that AI systems must be tailored to human needs in line with the commitment to put them to work in the service of humanity and the common good, seeking to enhance well-being and individual freedom, while at the same time maximising the benefits offered, while preventing and reducing the risks;

Amendment 76 #

Motion for a resolution
Paragraph 1 a (new)

1a. Asserts the principle whereby the development, spread and use of artificial intelligence, robotics and related technologies should always be at the service of human beings and never the other way round;

Amendment 77 #

Motion for a resolution
Paragraph 1 a (new)

1а. Stresses the utmost importance of implementing appropriate control measures, including as regards the adaptability, accuracy and explainability of the system concerned;

Amendment 78 #

Motion for a resolution
Paragraph 1 b (new)

1b. Is convinced that, in order to effectively exploit the advantages of AI, prevent possible misuse thereof and avoid regulatory fragmentation at EU level, it is essential to adopt legislation embodying principles adapted to future requirements throughout the EU and applicable to all AI systems; takes the view that, while sector-specific provisions are preferable for a wide range of possible applications, an effective and harmonised legal framework based on common principles and applicable across the board is necessary in order to establish equal standards throughout the Union and effectively protect our European values;

Amendment 79 #

Motion for a resolution
Paragraph 1 c (new)

1c. Takes the view that these common framework principles must be based on respect for all rights set out in international human rights law, the EU Treaties and the EU Charter of Fundamental Rights;

Amendment 8 #

Stéphane SÉJOURNÉ, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Recital A

A. whereas artificial intelligence, robotics and related technologies that have a lasting influence on all areas of activity throughout European society with the potential to directly impact all aspects of our societies, including basic social and economic principles and values, are being developed very quickly;

Amendment 80 #

Motion for a resolution
Paragraph 2

2. Considers that the determination of whether artificial intelligence, robotics and related technologies are to be considered high-risk as regards compliance with ethical principles should always follow from an impartial, regulated and external assessment;deleted

Amendment 81 #

Motion for a resolution
Paragraph 2

2. Considers that the determination of whether artificial intelligence, robotics and related technologies are to be considered high-risk ~~as regards compliance with ethical principles should always follow from an impartial, regulated and external assessment~~when, considering their intended use and the critical sectors where they are employed, their autonomous operation involves a significant potential to cause harm to one or more persons, in a manner that is random and impossible to predict in advance; considers that the significance of the risks depends on the interplay between the severity of possible harm, the likelihood that the risk materialises and the manner in which the technologies are being used;

Amendment 82 #

Motion for a resolution
Paragraph 2

Amendment 83 #

Motion for a resolution
Paragraph 2

Amendment 84 #

Motion for a resolution
Paragraph 2

2. Considers that the determination of whether artificial intelligence, robotics and related technologies are to be considered ~~high-risk as regards compliance with ethical principles~~to bear a risk should always follow from an impartial, regulated and external assessment;

Amendment 85 #

Motion for a resolution
Paragraph 2

2. Considers that the determination of whether artificial intelligence, robotics and related technologies are to be considered high-risk as regards compliance with ethical principles should ~~always follow from an impartial, regulated and external assessment~~be the subject of further discussion;

Amendment 86 #

Motion for a resolution
Paragraph 2 a (new)

2a. Underlines that the risk assessment of these technologies in light of ethical principles shall be complementary but not replace the risk assessment on the basis of objective criteria and in line with relevant sector- specific legislation applicable in different fields such, inter alia, as those of health, transport, employment, justice and home affairs, media, education and culture;

Amendment 87 #

Motion for a resolution
Paragraph 2 a (new)

2a. Considers that this assessment should be developed in a way that limits the administrative burden for companies, and SMEs in particular, as much as possible by using existing tools; notes that the Data Protection Impact Assessment list as provided for in Regulation (EU) 2016/679 could be one of those tools;

Amendment 88 #

Motion for a resolution
Paragraph 2 b (new)

2b. Considers that any artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies considered as high-risk from a compliance with ethical principles perspective shall not be classified as low- risk from a sector-specific legislation perspective;

Amendment 89 #

Motion for a resolution
Subheading 4

Safety ~~features~~, transparency ~~and~~, accountability, and accessibility

Amendment 9 #

Motion for a resolution
Recital А

А. whereas artificial intelligence, robotics and related technologies with the potential to directly impact all aspects of our societies, including basic rights and social and economic principles and values, are being promoted and developed very quickly;

Amendment 90 #

Motion for a resolution
Paragraph -3 a (new)

-3a. Recalls that the right to information of consumers is anchored as a key principle under the EU Treaties (Article 169 TFEU), and underlines that it therefore should be fully implemented in the artificial intelligence and automated decision-making context. It should especially encompass transparency regarding the fact that automation processes are involved and about their mode of functioning, such as how information is filtered and presented;

Amendment 91 #

Motion for a resolution
Paragraph -3 b (new)

-3b. Stresses that future regulation on artificial intelligence should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the artificial intelligence system, as well as the presumed opaqueness of decision- making; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no special legal obligations beyond those already in place; algorithmic systems that may harm an individual, violate rights and freedoms, affect an individual’s access to resources, or negatively impact their participation in society shall not be deemed to belong to the lowest risk category; this risk-based approach should be technologically neutral and follow transparent rules to establish legal certainty;

Amendment 92 #

Motion for a resolution
Paragraph -3 c (new)

-3c. Notes that it is essential for the risk assessment documentation, the software documentation, the algorithms and data sets used to be fully accessible to market surveillance authorities, while respecting Union law; additional prerogatives should be given to market surveillance authorities in this respect; stresses in this respect the importance of lawful reverse-engineering;

Amendment 93 #

Motion for a resolution
Paragraph -3 d (new)

-3d. Calls for a requirement for developers and deployers to provide for the relevant documentation on the use and design instructions, including source code, development tools and data used by the system, to be made easily accessible through a mandatory legal deposit, where a risk assessment so indicates; Notes that only such an obligation would allow authorities to assess the compliance of systems with Union law and ethical principles; Recommends that for vital and advanced medical appliances, these authorities, or other independent trusted entities, retain the means necessary to provide services, for example to persons carrying these appliances, such as maintenance, repairs, and enhancements, including software updates, especially in the case where such services are no longer carried out by the original supplier, to preserve human dignity, autonomy, and self-determination of the individual;

Amendment 94 #

Motion for a resolution
Paragraph 3

3. Maintains that artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies should be developed in a secure, traceable, technically rigorous ~~manner and in good faith~~, ethically and legally binding manner and in good faith; Considers especially that all players throughout the development and supply chains of artificial intelligence products and services should be legally accountable for the ethical aspects laid out in this Regulation;

Amendment 95 #

Stéphane SÉJOURNÉ, Karen MELCHIOR, Liesje SCHREINEMACHER, Adrián VÁZQUEZ LÁZARA

Motion for a resolution
Paragraph 3

3. Maintains that artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies, regardless of the field in which they are used, should be developed in a secure, dependable and technically rigorous manner and in good faith;

Amendment 96 #

Motion for a resolution
Paragraph 3

3. Maintains that ~~artificial intelligence, robotics and related~~high-risk technologies, including the software, algorithms and data used or produced by such technologies should be developed in a legal, secure, and technically r~~igor~~obust manner ~~and in good faith~~;

Amendment 97 #

Motion for a resolution
Paragraph 3

3. Maintains that high-risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies should be developed in a secure, technically rigorous manner ~~and in good faith~~;

Amendment 98 #

Motion for a resolution
Paragraph 3 a (new)

3а. Highlights the need for mechanisms to ensure liability and accountability in respect of AI systems and their outcomes;

Amendment 99 #

Motion for a resolution
Paragraph 4

4. Underlines that explainability is essential to ensuring that citizens trust these technologies, even if the degree of explainability is relative to the complexity of the technologies~~, and that~~; points out that it is not always possible to explain why a model has led to a particular result or decision, black box algorithms being a case in point; in such cases it should be complemented by auditability and traceability and transparent information regarding system capacities;

source: 652.548

2020/06/12 ENVI 173 amendments...

Amendment 1 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital A

A. Whereas the Union is founded on the ~~ethical~~ values stated in Article 2 of the Treaty on European Union;

Amendment 10 #

Draft opinion
Recital B

B. Wwhereas the artificial intelligence (AI) global leadership race~~, which will determine the source of ethical values and standards shaping the sector worldwide, is picking up the pace~~ is picking up the pace but should not be separated from ethical values and standards;

Amendment 100 #

Draft opinion
Paragraph 4 a (new)

4a. Stresses that an law-based Union AI ecosystem of trust, whether regarding the environment protection, health or food safety applications, extended by the Union AI ethical framework, will reinforce legal certainty and predictability, encourage stakeholders' involvement, increase the volume of entrusted data and market up take, allow for economies of scale and support an ecosystem of excellence in those sectors; is of the opinion that this will strengthen the Union AI sector's global competitiveness and the potential to promote Union values and standards;

Amendment 101 #

Draft opinion
Paragraph 4 a (new)

4a. Underlines the need to ensure that health data and data belonging to vulnerable groups such as persons with disabilities, patients, children, minorities and migrants are protected; stresses that the gathering, processing, sharing and access to data, must be in full respect of the established case-law of the European Court of Justice and of other European and international instruments that apply in the Union;

Amendment 102 #

Draft opinion
Paragraph 4 a (new)

4a. Recommends measures to encourage the involvement of all AI ethics stakeholders from the private sector, consumer groups and academia for the formulation of an ethical code tailored to technological, social and political developments;

Amendment 103 #

Draft opinion
Paragraph 4 a (new)

4a. Warns against attempts to give machines some kind of 'personality' which might result in the removal of human liability in the event of treatment errors;

Amendment 104 #

Draft opinion
Paragraph 4 b (new)

4b. Calls on the Commission to initiate an open, transparent sectoral dialogue giving priority to health care in order to then present an action plan to facilitate the development, testing and introduction of AI in research and innovation and the wide application of these in public health services;

Amendment 105 #

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including environmental, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services, patient-doctor relationship standards, data protection and privacy, civil liability in AI-assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 106 #

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including environmental, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services, patient-doctor relationship standards, interpretation of data protection and privacy legislation, including their interplay with research activities and drug development, civil liability in AI-assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 107 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ

Draft opinion
Paragraph 5

Amendment 108 #

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including in research, innovation and expertise in environmental, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services, patient-doctor relationship standards, data protection and privacy, civil liability in AI-assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 109 #

Sara CERDAS

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including environmental, public health, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services, patient-doctor relationship standards, data protection and privacy, civil liability in AI-assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 11 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital B a (new)

Ba. Whereas beyond EU action, many third countries are working on their ethical frameworks and there are multiple proposals at global level on AI principles among which the Principles on AI2a signed by OECD members in May 2019, the G20 Ministerial Statement of 2019 on Trade and Digital Economy whose annex contains the principles for AI, the 2019 Council of Europe “Guidelines on Artificial Intelligence and Data Protection”2b and the IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems2c; _________________ 2a https://legalinstruments.oecd.org/api/print ?ids=648⟨=en 2bhttps://rm.coe.int/guidelines-on- artificial-intelligence-and-data- protection/168091f9d8 2c https://ethicsinaction.ieee.org

Amendment 110 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including environmental, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services and technologies, patient-doctor relationship standards, data protection and privacy, civil liability in AI- assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 111 #

Draft opinion
Paragraph 5

5. Strongly supports the Commission in establishing a minimal common Union AI ethical framework to counter the shortcomings caused by AI internal market fragmentation, including environmental, healthcare, and food safety applications, and to prevent AI double standards across Member States for AI developed in Union and beyond, inter alia in areas such as consumer data management, protection and privacy in smart grids, waste management, equal access to services, patient-doctor relationship standards, data protection and privacy, civil liability in AI-assisted public healthcare, civil liability regarding autonomous vehicles or machinery;

Amendment 112 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ, Frédérique RIES

Draft opinion
Paragraph 5 a (new)

5a. Stresses that the health sector has specific characteristics; agrees with the Commission that citizens should have secure access to a comprehensive electronic record of their health data, and should remain in control of and be able to share their personal health data securely with authorised parties, while, in accordance with the data protection legislation, unauthorised access should be prohibited; notes, further, that data should be stored on secure local servers and processed by independent bodies;

Amendment 113 #

Draft opinion
Paragraph 5 a (new)

5a. Strongly supports the creation of a European Health Data Space1a proposed by the Commission which aims at promoting health-data exchange and at supporting research in full respect of data protection, including processing data with AI technology, and which strengthens and extends the use and re-use of health data; encourages the upscaling of cross-border exchange of health data, their link and use through secure, federated repositories, specific kinds of health information, such as European Health Records (EHRs), genomic information, and digital health images to facilitate Union-wide interoperable registers or databases in areas such as research, science and health sectors; _________________ 1a Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions - A European strategy for data, COM(2020)0066

Amendment 114 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ

Draft opinion
Paragraph 5 a (new)

5a. Underlines the importance of training highly skilled professionals in environmental, health and food safety- related areas that should go hand-in-hand with the need of having diverse teams of developers and engineers working alongside key actors to prevent gender and cultural bias of being inadvertently included in AI algorithms, systems and applications. The mutual recognition of such qualifications across the Union should be also ensured;

Amendment 115 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 5 a (new)

5a. Considers that in areas such as health, liability must ultimately lie with a natural or legal person; emphasizes the need for traceable and publicly available training data for algorithms;

Amendment 116 #

Draft opinion
Paragraph 5 b (new)

5b. Calls for the right balance between privacy and data protection and data utility; considers that it is important for scientific advancement to ensure the ability to share and process health data in sufficient depth and detail; ensures data anonymization while avoiding excessive data minimization; calls for interoperable, suitable databases, registers and repositories at EU level to facilitate their use in health, environment and food safety sectors;

Amendment 117 #

Draft opinion
Paragraph 5 b (new)

5b. Stresses that neither insurance companies nor any other type of service provider should be authorised to use e- health data to introduce discrimination in the setting of prices, given that this would run counter to the fundamental right to the highest attainable standard of health;

Amendment 118 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 5 b (new)

5b. Emphasises that patients should know when and how they are interacting with a human professional and when they are not; insists that patients should have the freedom to decide about this interaction and should be offered an alternative of equal standard;

Amendment 119 #

Draft opinion
Paragraph 5 c (new)

5c. Points out that, to take decisions, robots use algorithms which play the part of values and ethical frameworks, and that their introduction has significant ethical implications for healthcare and social relations; is particularly concerned about the use for paedophilic and sexual abuse purposes of companion robots; believes that ethical considerations should be taken into account in the design of robotics technologies; calls, in the development process for these machines, for a place to be granted to ethics, based on an approach such as value-sensitive design, particularly with regard to care robots for the elderly and companion robots; stresses that this approach should also be adjusted to take account of animal welfare;

Amendment 12 #

Draft opinion
Recital B a (new)

Ba. Whereas the main AI research and development hubs are the United States, the EU and China, which alone account for three quarters of AI actors and, respectively, 28%, 25% and 23% of resources;

Amendment 120 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK

Draft opinion
Paragraph 6

6. Calls for securing the full enforcement of Union legal framework on data protection and privacy, relevant notably in the healthcare AI applications and related sensitive data, to strengthen the “Right to an explanation” foreseen in Article 22 of Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation, (GDPR))2 ~~and higher interpretability requirements for high-risk AI;~~ ; encourages the Commission during the GDPR review to specifically assess whether Union data protection legislation is fit for purpose or whether additional regulatory safeguards are required, such as: rules on harmonized data anonymization techniques for health data; rules on data access and data control when it comes to the use of algorithm-based solutions/automated decisions and multiple-source data; quality and safety standards for all information systems where health data is processed, as well as the mechanism to monitor compliance of all parties involved in the processing of personal biomedical data with privacy protection laws, product liability, transparency, accountability and other ethical norms; _________________ 2Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).

Amendment 121 #

Draft opinion
Paragraph 6

6. Calls for securing the compliance and full enforcement of Union legal framework on data protection and privacy~~, relevant notab~~ and ensuring that personal data is protected, especially in the healthcare AI applications and ~~related~~other sensitive data, to strengthen the “Right to an explanation” foreseen in Article 22 of Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation, (GDPR))2 and higher interpretability requirements for high-risk AI; reminds of the risk of malicious data alterations and manipulation, and possible hacks or data thefts, can be particularly severe in health sector and can be used to harm, discredit or profit from individual; _________________ 2Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).

Amendment 122 #

Draft opinion
Paragraph 6

Amendment 123 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 6

6. Calls for citizen and patient empowerment regarding their personal data and for securing the full enforcement of Union legal framework on data protection and privacy, relevant notably in the healthcare AI applications and related sensitive data, to strengthen the “Right to an explanation” foreseen in Article 22 of Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation, (GDPR))2 and higher interpretability requirements for high-risk AI; _________________ 2Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).

Amendment 124 #

Draft opinion
Paragraph 6

6. Calls for securing the full enforcement and a uniform interpretation of Union legal framework on data protection and privacy, relevant notably in the healthcare AI applications and related sensitive data, to strengthen the “Right to an explanation” foreseen in Article 22 of Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation, (GDPR))2 and higher interpretability requirements for high-risk AI; _________________ 2Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).

Amendment 125 #

Draft opinion
Paragraph 6 a (new)

6a. Emphasises the need to ensure that AI-driven medical devices should comply with the safety and performance requirements of the European Medical Devices Regulation; calls on the Commission and Member States to ensure that the Medical Devices Regulation is implemented with a view to these technologies; considers new guidelines and specifications are required for the evaluation of safety and effectiveness of software, AI and deep-learning powered devices throughout the entire usage cycle;

Amendment 126 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ, Frédérique RIES

Draft opinion
Paragraph 6 b (new)

6b. Underlines that, in addition to clear regulatory requirements on accountability and liability, there is also a need to ensure algorithmic transparency, so that it is possible to track the moment when 'things went wrong' and allow for the timely intervention by experts; considers algorithmic transparency as crucial to prevent situations where medical decision-making is done in a 'black-box' environment; underlines that black-box algorithms that make inexplicable decisions are unacceptable in any sector but in a context where AI decision-making has an impact on life or death decisions, the consequences of algorithmic failure could be grave; calls on the Commission and Members States to open dialogue with key stakeholders from the fields of medicine, IT, mathematics, physics, medical data technology, clinical psychology, bioengineering and pharmaceutical to establish dialogue-building platforms and assess the impact on the doctor-patient relationship and the dehumanisation of medical care.

Amendment 127 #

Draft opinion
Paragraph 6 a (new)

6a. Considers that there are risks of biases and discrimination in the development, deployment and use of high- risk artificial intelligence, robotics and related technologies, including the software, algorithms and data used or produced by such technologies; recalls that, in all circumstances, those technologies should respect human dignity and ensure equal treatment for all; considers that such possible bias could be addressed by setting rules on data processing and setting up appropriate safeguards against bias and discrimination based on social, economic, ethnic, racial, sexual, gender, disability or other factors; warns of potential misuse of AI diagnostic applications and calls for AI capability and motivational safeguards;

Amendment 128 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 6 a (new)

6a. Highlights, in the healthcare sector, that data originated from patients using AI technologies should fulfil all the privacy requirements as stipulated in the GDPR, and that by no means, the data generated should contribute to any kind of discrimination (known or novel); calls on the Commission and Member States to guarantee that data accessibility to private companies, such as health or life insurance companies, is prevented and that the “right to be forgotten” of patients is fully respected;

Amendment 129 #

Draft opinion
Paragraph 6 a (new)

6a. Stresses the connection between reflection on the AI ethical framework and the European data strategy; calls for the introduction of a 'data of public interest' or 'data of general interest' concept for private data that could be opened up for the purposes of a public research goal (energy production and consumption, biodiversity data, climate data, etc.), in line with work by the Internet Governance Forum;

Amendment 13 #

Draft opinion
Recital B a (new)

Ba. whereas the EU is currently lagging behind Asia and North America in terms of AI investment and patents;

Amendment 130 #

Draft opinion
Paragraph 6 a (new)

6a. Stresses that the issue of opacity of the AI “black box-effect” machine deep learning, its complexity and partially autonomous behaviour might prevent the enforcement of the Union acquis, including its ethical values and standards and undermine the Union AI ecosystem of trust among investors and consumers, and thus hamper the AI ecosystem of excellence of the Union;

Amendment 131 #

Draft opinion
Paragraph 6 a (new)

6a. Calls on the Commission to promote and fund the development of human-centric artificial intelligence, robotics and related technologies that address environment and climate challenges and that ensure equal access to and enjoyment of fundamental rights through the use of tax, green public procurement, or other incentives;

Amendment 132 #

Draft opinion
Paragraph 6 a (new)

6a. Believes that users should have full control over their data, which should not be used to harm or discriminate against them; AI systems should therefore be designed with guaranteed confidentiality and data protection in mind;

Amendment 133 #

Draft opinion
Paragraph 6 a (new)

6a. Calls on the Member States to align medical education and training programmes to developments in AI, as well as an awareness of the ethical challenges associated with the application of AI;

Amendment 134 #

Draft opinion
Paragraph 6 b (new)

6b. Stresses that algorithms and machines that have the capacity to learn and evolve should be designed in a transparent and respectful manner from the outset, with particular attention given to their implications for workers' and consumers' physical and mental well- being; stresses that workers should be able to benefit from a right to an explanation on decisions taken by algorithms with a view to reducing uncertainty and opacity, which are harmful to workers' well-being in the long term; calls on the Commission and Member States to legislate to that effect; urges the Commission to put forward a legislative initiative regarding platform micro-taskers to provide them with legal protections, which are essential to their physical and mental well-being;

Amendment 135 #

Draft opinion
Paragraph 6 b (new)

6b. Calls for Union guiding initiatives promoting interpretable algorithms, eXplainable AI (xAI), symbolic reasoning AI, white box AI testing technics, by showing that those technologies can be combined with deep neural networks and by showing its legal, ethical and often business advantages, and also promoting methods to determine risks connected with different technological options using among others the experience of the UK’s Information Commissioner's Office (ICO) and The Alan Turing Institute guidelines “Explaining decisions made with AI”, showing that even highly complex neural AI systems can be interpreted sufficiently;

Amendment 136 #

Draft opinion
Paragraph 6 c (new)

6c. Reiterates that opportunities and risks inherent to these technologies have a global dimension that requires a consistent harmonised approach at international level; calls on the Commission to work in bilateral and multilateral settings to advocate and ensure ethical compliance;

Amendment 137 #

Draft opinion
Paragraph 6 c (new)

6c. Welcomes a European strategy for data, addressing challenges ahead for the Union in this area that is key to AI progress, and seeking European opportunities for competitive advantage in new data economy, especially in the growing sector of decentralised, non- personal data coming from industry, business and the public sector and from devices at the edge of the network, which is expected to constitute 80% of 175 zettabytes in 2025 and reverse current proportions;

Amendment 138 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK

Draft opinion
Paragraph 6 d (new)

6d. Stresses that a data economy fuelling AI development, which is called therefore the crude oil of 21st century, is not without risks; underlines that AI ethical and thus legal challenges extend beyond data protection and privacy, to liability for data environment used for AI training, testing and, in some cases, to data input used for algorithm self- adaptation during functioning, which may result in bias, discrimination or mistakes; welcomes requirements proposed in the White Paper for high risk AI training data, addressing as well safety – sufficiently broad data to cover all relevant scenarios in order to avoid dangerous situations as discrimination - sufficiently representative data to reflect well the social environment it will be applied to;

Amendment 139 #

Draft opinion
Paragraph 7

7. Supports the view that the seven AI requirements identified in the Ethics Guidelines for Trustworthy AI of the High- Level Expert Group on AI constitute solid building blocks for a common Union AI ethical framework, addressing, among others, ethical aspects of AI applications in environment, health and food protection; calls for an improvement of the acquis on transparency, traceability and human oversight, which were indicated as areas in need of further improvement in the feedback given on the Guidelines by 350 organisations; furthermore, encourages the creation of the Union AI ethical framework in a spirit of openness to the works of other international partners that share Union values~~, e.g. to the Rome Call for AI Ethics by Pope Francis~~;

Amendment 14 #

Draft opinion
Recital C

C. Whereas AI solutions may benefit society in the areas of green transition, environment protection, waste management, climate change, energy management and efficiency, air quality e.g. smart grids and electro-mobility; whereas, however, it is essential to minimise the ecological footprint of robotics, as the use of cyber-physical systems and robots is expected to increase overall energy consumption and the amount of electrical and electronic waste; whereas, lastly, 14% of current jobs in the OECD area could disappear over the next 15 to 20 years because of automation, with 32% undergoing radical change;

Amendment 140 #

Draft opinion
Paragraph 7

7. Supports the view that the seven AI requirements identified in the Ethics Guidelines for Trustworthy AI of the High- Level Expert Group on AI constitute solid building blocks for a common Union AI ethical framework, addressing, among others, ethical aspects of AI applications in environment, health and food protection; calls for an improvement of the acquis on transparency, traceability and human oversight, which were indicated as areas in need of further improvement in the feedback given on the Guidelines by 350 organisations; furthermore, encourages the creation of the Union AI ethical framework in a spirit of openness to the works of other international partners that share Union values, ~~e.g. to the Rome Call for AI Ethics by Pope Francis~~particularly the UN and the Council of Europe, which has drawn up a European ethical charter on the use of artificial intelligence in judicial systems and has its own legal research centre, the Ad Hoc Committee on Artificial Intelligence (CAHAI);

Amendment 141 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ, Frédérique RIES

Draft opinion
Paragraph 7

7. Supports the view that the seven AI requirements identified in the Ethics Guidelines for Trustworthy AI of the High- Level Expert Group on AI constitute solid building blocks for a common Union AI ethical framework, with proper legal anchoring, addressing, among others, ethical aspects of AI applications in environment, health and food protection; calls for an improvement of the acquis on transparency, traceability and human oversight, which were indicated as areas in need of further improvement in the feedback given on the Guidelines by 350 organisations; furthermore, encourages the creation of the Union AI ethical framework in a spirit of openness to the works of other international partners that share Union values, e.g. to the Rome Call for AI Ethics by Pope Francis;

Amendment 142 #

Draft opinion
Paragraph 7 a (new)

7a. Calls to ensure transparency, responsibility, auditability, predictability and accountability, as citizens, patients and users should be informed when interacting with a system using artificial intelligence by clear and understandable explanations of the data used, of the functioning of the algorithm, of its purpose, of its outcomes, and of its potential dangers; underlines that transparency and explainability are essential to ensure trust in these technologies; considers that the explanation should be complemented by auditability and traceability as respect to such principles is a guarantee to accountability; reminds that AI applications can outperform the humans at narrow specific tasks while failing in overview analysis; calls for human oversight, professional responsibility and system predictability with ability to override the AI system;

Amendment 143 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ, Frédérique RIES

Draft opinion
Paragraph 7 a (new)

7a. Supports the recommendation of the High-Level Expert Group on AI to establish 720 expert chairs for AI ethics, 'one for every major European university'; considers that experts appointed to these positions could also strengthen links between science and civil society by convening public debates, interpreting research and offering interdisciplinary input to a new generation of students;

Amendment 144 #

7b. Considers that any natural or legal person should be able to seek redress for a decision issued by a high-risk artificial intelligence, robotics or related technology at his or her detriment and that any decision taken by AI should be subject to strict human verification and due process; suggests that safeguards related to the use of high-risk artificial intelligence, robotics and related technologies within the framework of public power decisions including periodic assessment and possible review of regulatory framework to keep up with technological development, establishing binding guidelines on the methodology of the compliance assessment to be followed by the national supervisory authorities, and establishing non-binding guidelines directed to the developers, the deployers and the users;

Amendment 145 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for disease prevention and control~~, exemplified by AI predict~~ and climate change monitoring, particularly ing the case of zoonoses, such as COVID-19 ~~epidemic before WHO; urges the Commission to equip ECDC in its reform, which was brought about by~~, whose development and spread is exacerbated by deforestation; stresses that the development and use of AI tracking technologies to respond to the COVID-19 ~~crisis, with the legal framework and resources allowing for gathering necessary data independently, including, among others, AI solution~~pandemic has raised a number of ethical questions on users' rights; stresses, further, that some tracing applications were clearly violating the EU's ethical values;

Amendment 146 #

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for disease prevention and control, exemplified by AI predicting the COVID19 epidemic before WHO; urges the Commission to adequately equip ECDC in witsh re~~form, which was brought about by the COVID19 crisis, w~~sources allowing for ith t~~he legal framework and resources allowing for~~o utilize AI solutions and gather~~ing~~ necessary data independently, ~~including, among others, AI solution~~to address issues revealed by the COVID19 crisis;

Amendment 147 #

Sara CERDAS

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for improved health, disease prevention and control, exemplified by AI predicting ~~the COVID19 epidemic before~~public health threats where early warning can be given to the authorities such as the ECDC and WHO; urges the Commission to equip ECDC in its reform, which was brought about by the COVID-19 crisis, with the legal framework and resources allowing for gathering necessary data independently in conjunction with the Member States, including, among others, AI solutions;

Amendment 148 #

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for disease prevention and control, exemplified by AI predicting the COVID19 epidemic before WHO; urges the Commission to equip ECDC in its reform, which was brought about by the COVID19 crisis, with the legal framework and resources allowing for gathering necessary data independently and in compliance with data protection and privacy, including, among others, AI solutions;

Amendment 149 #

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for disease prevention, treatment and control, exemplified by AI predicting the COVID19 epidemic before WHO; urges the Commission to equip ECDC in its reform, which was brought about by the COVID19 crisis, as well as EMA, with the legal framework and resources allowing for gathering necessary data independently, including, among others, AI solutions;

Amendment 15 #

Draft opinion
Recital C

C. Whereas AI solutions and other emerging digital technologies may benefit society in the areas of green transition, environment protection, waste management, climate change, energy management and efficiency, air quality e.g. smart grids and electro-mobility; whereas the use of algorithmic systems for surveillance and behavioural regulation represents a serious threat to European values and algorithmic transparency;

Amendment 150 #

Draft opinion
Paragraph 8

8. Highlights the benefits of AI for disease prevention and control, exemplified by AI predicting the COVID19 epidemic before WHO; urges the Commission to equip ECDC in its reform, which was brought about by the COVID19 crisis, with the legal framework and resources allowing for gathering necessary anonymized data independently, including, among others, AI solutions;

Amendment 151 #

Draft opinion
Paragraph 8 a (new)

8a. Stresses that the development of AI applications might bring down the costs and increase the volume of services available, e.g. health services, public transport, Farming 2.0, making them more affordable to a wider spectrum of society; stresses that AI applications may also result in the rise of unemployment, pressure on social care systems, increase of poverty; emphasizes in accordance with the values enshrined in Article 3 of the Treaty on European Union the need to adapt Union AI transformation to socio- economic capacities, adequate social shielding, education and creation of alternative jobs; calls to consider the establishment of a Union AI Adjustment Fund building upon the experience of The European Globalisation Adjustment Fund (EGF) or the currently developed Just Transition Fund;

Amendment 152 #

Draft opinion
Paragraph 8 a (new)

8a. Calls for the EU not to slip into 'technology solutionism' and to maintain a human role in decisions and public policies; considers that, particularly in the health sector, it is useful for robots to be present to support the work of doctors or healthcare assistants, with a view to improving the human experience of diagnosis and treatment, without, however, disregarding the need to ensure that medical practice and patient care practices are not dehumanised;

Amendment 153 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 8 a (new)

8a. Points out that the use of tracking and contact tracing technologies by public authorities during the COVID 19 crisis and other potential health emergencies might conflict with data protection; recalls in this regard the Commission Guidance on applications supporting the fight against the COVID 19 pandemic in relation to data protection and the need for proportionality, limitation in time, alignment with European values and respect of human dignity and fundamental rights;

Amendment 154 #

Draft opinion
Paragraph 8 a (new)

8a. Calls for the creation of shared European databases analysed by AI in healthcare for using anonymised medical data for faster, earlier and more precise diagnosis, as AI and robotics bring massive opportunities and can process enormous amounts of data in no time; considers that these systems need enough comparable data to function efficiently and at the same time they need to be GDPR compliant and protect personal data;

Amendment 155 #

8a. Notes that given the increasing development of AI applications, which require computational, storage and energy resources, the environmental impact of AI systems should be considered through their lifecycle;

Amendment 156 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Paragraph 8 b (new)

8b. Highlights that AI and robotics can bring numerous benefits to our environment, health and food safety as the dematerialisation of the economy makes the Union less dependent on raw materials or on the increased use of personalised medicine; underlines however, that their carbon footprint remains still high; calls on the Commission to carry out a study on the impact of AI technology’s carbon footprint and the positive and negative impacts of the transition to the use of AI technology by consumers; further calls on the Commission to include the footprint information in the common European Dataspace for Smart Circular Applications foreseen in the EU Action Plan on Circular Economy and to deal specifically with these technologies within the ICT key value chain of the above- mentioned plan;

Amendment 157 #

Draft opinion
Paragraph 8 b (new)

8b. Calls for a Union standardised inter-operability of eHealth applications and a creation of common European data access for prescriptions, diagnosis and medical reports, simply accessible for all Union citizens and in all Member States;

Amendment 158 #

Draft opinion
Paragraph 8 b (new)

8b. Calls for the education of professionals in the area of environment protection, health and food safety preparing for AI applications and rising awareness of AI risks and ethical challenges;

Amendment 159 #

Irena JOVEVA, Véronique TRILLET-LENOIR, Nils TORVALDS, Susana SOLÍS PÉREZ, Martin HOJSÍK

Draft opinion
Paragraph 8 c (new)

8c. Considers that AI and robotics can provide considerable improvements in the control of medical devices and facilitate the everyday work of health professionals; considers that for critical medical devices, there needs to be a back-up system in place to monitor and secure the functionality of the device in any possible situation of interference and that possible cyber threats in the control of such devices need to be taken into consideration and mitigated; stresses that apart from hackers and outside threats, cyber threats can also originate from human mistakes or system errors and that it is necessary to have adequate back-up systems in place and operational; considers furthermore that the Union should create an AI backup development roadmap to address the possible issues of AI system controls making an error;

Amendment 16 #

Draft opinion
Recital C

C. Whereas AI solutions may benefit society ~~in the areas of green transition, environment protection, waste management, climate change, energy~~and contribute to the development of better strategies and innovations in the areas of green transition, environment protection by increasing the efficiency of agriculture, waste management, mitigation and adaptation of climate change, greening of various industrial processes, energy and transport management and efficiency, air quality e.g. smart grids and electro-mobility;

Amendment 160 #

Draft opinion
Paragraph 8 c (new)

8c. Stresses that the public sector should focus on solving social problems rather than generating AI uptake for its own sake; calls for the improvement of the public procurement regulations and guidelines of the Union, including EU Green Public Procurement, so that during relevant evaluation procedures for tender offers, one takes into account whether a given issue requires an AI system application, and allows to follow an alternative delivery path in cases where the evaluation indicates that such a non- AI solution addresses the social problem better;

Amendment 161 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK

Draft opinion
Paragraph 8 d (new)

8d. Considers it important to make sure no one is left behind in an AI system or a domain design, especially in consumer use; stresses that systems should be user-centric and AI products and services designed accessible for all people to use regardless of their age, gender, abilities or characteristics and that AI systems should not have a one- size-fits-all approach and should consider Universal Design principles addressing the widest possible range of users, following relevant accessibility standards;

Amendment 162 #

Draft opinion
Paragraph 9

9. Calls for securing sufficient financing for the Union AI transformation; supports the ambitions laid out in the Commission White Paper to attract €200 billion of AI public and private investment in the next 10 years in the Union; welcomes the attention granted to deficits of AI ecosystems in less-developed regions and to the needs of SMEs and start-ups; calls on the Commission to facilitate geographically balanced access to all AI funding, including for SMEs and start-ups; ~~stresses that the new Union objectives must not diminish Union engagement in its long standing priorities, like the CAP or Cohesion Policy.~~

Amendment 163 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Paragraph 9

9. Calls for securing sufficient financing for the Union AI transformation; supports the ambitions laid out in the Commission White Paper to attract €200 billion of AI public and private investment in the next 10 years in the Union; welcomes the attention granted to deficits of AI ecosystems in less-developed regions and to the needs of SMEs and start-ups; calls on the Commission to ~~facilitate geographically balanced access to all~~identify public infrastructure and promote the prioritization of AI funding, in cl~~uding for SMEs and start- ups~~imate change mitigation and adaptation, renewable energies and health; stresses that the new Union objectives must not diminish Union engagement in its ~~long~~ standing priorities, like the CAP or, Cohesion Policy., the Green Deal and the COVID19 Recovery Plan;

Amendment 164 #

Draft opinion
Paragraph 9 a (new)

9a. Criticises the fact that the development of digital farming promised by the Commission as part of the CAP reforms consolidates the stranglehold of an industrial farming system that is very harmful for the environment and biodiversity; calls for digital technologies to serve to develop a European farming model based on agro-ecology; calls for preference to be given to open-source platforms, run directly by farmers, providing mutual assistance and sharing good practices among farmers, and putting producers of foodstuffs in direct contact with consumers through short supply chains; insists that farmers must keep ownership and control over data produced on their holdings, even after aggregation;

Amendment 165 #

Draft opinion
Paragraph 9 a (new)

9a. Stresses that the development of AI and related technologies raises questions regarding data storage and the resulting pollution; recalls, in that regard, the high energy cost of storing data in data centres, which alone account for 4% of global electricity consumption; calls, in that connection and in parallel with initiatives taken by GAFAM, on the EU to consider establishing European 'green data centres', which would enable the EU to guarantee its independence in the collection and management of data and, at the same time, ensure that this storage is ethical and sustainable;

Amendment 166 #

Draft opinion
Paragraph 9 a (new)

9a. Calls for securing a clearer legal remit and sufficient financing for EMA and national competent authorities for medicines to support innovation and public health aspects related to AI in the medicine lifecycle, in particular to collect and analyse real world health data that can generate additional evidence on medicinal products to support R&D and to optimise the safe and effective use of existing medicines in the interest of patients and of the European healthcare systems;

Amendment 167 #

Draft opinion
Paragraph 9 b (new)

9b. Recalls, further, that data centres, which are essential to the development of appropriate AI, are known to be a major source of heat loss, and this has an impact on the environment; proposes, within the context of ethical reflections on the development of AI, that an ethical charter for businesses be drawn up with a view to establishing a circular economy on the reuse of heat produced by data centres;

Amendment 168 #

Draft opinion
Paragraph 9 b (new)

9b. Calls on the Commission to establish a temporary ban on buying European AI and robotics industry businesses that are currently undervalued or facing commercial troubles as a result of the coronavirus crisis;

Amendment 169 #

Draft opinion
Paragraph 9 c (new)

9c. Points out that an AI ethical framework must not be limited to simply data and data use, but must also cover conditions for the production of AI technologies; recalls, in this connection, that the requisite raw materials and production chains are mainly in third countries, in which labour conditions and treatment of workers do not meet the EU's ethical standards;

Amendment 17 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Recital C

C. Whereas AI solutions may benefit society in the areas of green transition, environment ~~protection,~~and biodiversity protection, circular economy and waste management, climate change, energy management and efficiency, water and air quality, e~~.g. smart grids and electro-mobility~~arth observation and risk management, among others;

Amendment 170 #

Draft opinion
Paragraph 9 c (new)

9c. Supports the Commission in the establishment of a list of assets and businesses in 'critical' domains, such as health, medical research, biotechnology, digital, artificial intelligence, robotics and infrastructure essential to our security and to public order; calls for a ceiling to be set to prevent foreign businesses in critical domains from acquiring majority stakes;

Amendment 171 #

Draft opinion
Paragraph 9 d (new)

9d. Recognises the major role that AI and digital technologies may play with regard to human health; welcomes, in this connection, the major developments in genomics, radiology (particularly cancer detection), dermatology and ophthalmology;

Amendment 172 #

Draft opinion
Paragraph 9 e (new)

9e. Acknowledges the advantages of monitoring health via a mobile application, including for individual pathologies such as addiction, insomnia, depression and pathologies that require more intense monitoring, such as epilepsy, Parkinson's disease and multiple sclerosis;

Amendment 173 #

Draft opinion
Paragraph 9 f (new)

9f. Recalls, however, AI's current limits, and that although constant progress is being made with AI, it is no replacement for a doctor's expertise or human contact; recalls that at this stage of our knowledge of AI programming, individuals remain responsible for how they programme algorithms and the information that they feed into an application; calls on the Commission, in the context of this framework of ethical aspects of artificial intelligence, robotics and related technologies, to include the fundamental principle that AI must remain a tool that serves professionals and the general public, and for its legal status to reflect this principle clearly;

Amendment 18 #

Draft opinion
Recital C

C. Whereas AI solutions may benefit society in the areas of green transition, environment protection, waste management, circular economy, climate change, energy management and efficiency, air quality e.g. smart grids and electro-mobility;

Amendment 19 #

Draft opinion
Recital C a (new)

Ca. Whereas the number of those who will be durably affected by the deployment of AI is huge; whereas, in this connection, Microsoft journalists were recently laid off and replaced by AI carrying out the same content-sorting task; whereas, in addition, AI has regularly pushed articles from questionable information sources and indeed fake news;

Amendment 2 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Frédérique RIES, Martin HOJSÍK, Susana SOLÍS PÉREZ

Draft opinion
Recital A

A. Whereas the Union is founded on the ethical values stated in Article 2 of the Treaty on European Union and on compliance with the precautionary principle stated in Article 191(2) of the Treaty on the Functioning of the European Union;

Amendment 20 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital C a (new)

Ca. Whereas in its digital package published on 19 February 2020 the Commission states that ICT today accounts for between 5% and 9% of global electricity consumption and 2% of CO2 emissions and that the volume of data transferred and stored will continue to grow exponentially in the years to come; whereas the 2018 Joint Research Centre study “Artificial Intelligence/A European Perspective” estimates that data centres and data transmission could account for 3-4% of all power consumption of the Union;

Amendment 21 #

Draft opinion
Recital C a (new)

Ca. Whereas AI solutions can contribute to reduction of the current environmental footprint of the ICT sector which is estimated at more than 2% of all global emissions; whereas the European digital strategy proposes green transformation measures for digital sectors;

Amendment 22 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Susana SOLÍS PÉREZ, Frédérique RIES

Draft opinion
Recital C b (new)

Cb. having regard to AI energy performance; whereas a standard machine-learning project now produces some 284 tonnes of carbon dioxide equivalent throughout its development cycle, which is five times more than the emissions from a car from when it is manufactured until it is scrapped; whereas simultaneously pushing the Internet of Things and action to combat the climate crisis is therefore absurd and societies must bring about major change in production and consumption;

Amendment 23 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital C b (new)

Cb. Whereas private and independent research exists which warns of the high carbon footprint of deep learning and AI3a and the need to find solutions in this regard; _________________ 3a https://www.technologyreview.com/2019/0 6/06/239031/training-a-single-ai-model- can-emit-as-much-carbon-as-five-cars-in- their-lifetimes/

Amendment 24 #

Draft opinion
Recital C b (new)

Cb. Whereas the development of artificial intelligence, robotics, automated decisions, machine learning, and related technologies are a critical enabler for attaining the goals of the Green Deal and to reach the sustainability goals of the Green Deal in many different sectors; whereas digital technologies can boost the impact of policies in delivering environmental protection;

Amendment 25 #

Draft opinion
Recital C c (new)

Cc. Whereas guideline 6 in the Commission communication 'Building Trust in Human-Centric Artificial Intelligence', based on the work of the group of experts, focuses on sustainability and ecological responsibility of AI systems; whereas acknowledgement of the finite nature of Earth's resources is an ethical stance in terms of Union policy, as demonstrated by the Green Deal or the declaration of a state of climate and environmental emergency; whereas the AI ethical framework must therefore encompass ecodesign and component recyclability criteria;

Amendment 26 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital C c (new)

Cc. Whereas properly regulated AI would help guide efforts to achieve the UN SDGs and help reach the climate objectives of the Paris Agreement;

Amendment 27 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK, Susana SOLÍS PÉREZ

Draft opinion
Recital D

D. Whereas AI can be applied to almost any field in medicine: biomedical research, exemplified by the AI-discovered antibiotic Halicin or AI contributions to new cancer therapies, medical education, clinical decision-making, personalizsed medicine, psychiatric diagnosis and treatment, in revolutionizsing robotic prostheses and support systems, telemedicine and the overall efficiency of the health systems; whereas this requires appropriate training and preparation for health and administrative personnel; whereas digital health should not be a gateway to dehumanised care; whereas, in addition, there is a considerable risk that the digital divide, if not resolved, will be instrumental in the spread of 'medical deserts' across Europe;

Amendment 28 #

Draft opinion
Recital D

D. Whereas AI can be applied to almost any field in medicine: biomedical research, exemplified by the AI-discovered antibiotic Halicin or AI contributions to new cancer therapies, ~~medical education~~help in diagnosis, earlier disease detention and prevention with methods such as predicative medicine and/or identification of risk factors, medical education, assisting caregivers, supporting elderly care, monitoring patient conditions, more efficient development of medicines, more targeted treatment, clinical decision- making, personalized medicine, psychiatric diagnosis and treatment, in revolutionizing robotic prostheses and support systems, telemedicine and the overall efficiency and interoperability of the health systems;

Amendment 29 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Recital D

D. Whereas AI can be applied to almost any field in medicine: biomedical research, exemplified by the AI-discovered antibiotic Halicin or AI contributions to ~~new~~ cancer prevention, more precise diagnosis and new therapies, medical education, clinical decision-making, personalized medicine, psychiatric diagnosis and treatment, in revolutionizing robotic prostheses and support systems, telemedicine, telesurgery and the overall efficiency of the health systems;

Amendment 3 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital A a (new)

Aa. Whereas rapid advances in contemporary scientific research and innovation in the fields of environment, health and food safety have raised a number of important ethical, legal and social issues that affect the relationship between science and society;

Amendment 30 #

Draft opinion
Recital D

Amendment 31 #

Draft opinion
Recital D a (new)

Da. Whereas personal health data must remain a patient's property; whereas no information concerning the health of patients must be passed on without their full and informed consent; whereas personal health data, which by their nature are extremely sensitive, must be scrupulously safeguarded; whereas the highest cybersecurity standards ought to be established for the relevant networks; having regard in particular to the interesting initiatives emerging from a number of AI developers during the crisis that offer hospitals anti-spam solutions to safeguard their e-mail systems free of change, and subject to no obligation, against cyberattacks;

Amendment 32 #

Draft opinion
Recital D a (new)

Da. Whereas the application of AI, robotics and related technologies should not diminish or harm the doctor-patient relationship, but should provide doctors with assistance in diagnosing and/or treating patients more effectively;

Amendment 33 #

Draft opinion
Recital D b (new)

Db. whereas there are serious ethical concerns about the autonomy of machines and their influence on the doctor-patient relationship;

Amendment 34 #

Draft opinion
Recital D a (new)

Da. whereas AI technology will accelerate the digital transformation of industry and play an essential part in the success of the digital economy in an increasingly connected world;

Amendment 35 #

Draft opinion
Recital D a (new)

Da. whereas AI will play an increasingly significant role, bearing in mind our ageing societies and possible shortages of healthcare professionals;

Amendment 36 #

Draft opinion
Recital E

E. Whereas current policy and ethical guidelines for AI are lagging behind ethical challenges that must be identified and mitigated, since AI has tremendous capability to threaten patient preference, safety, and privacy; whereas the boundaries between the roles of physicians and machines in patient care need to be outlined, as regards AI and digital, the EU favours a soft digital ethics approach, prioritising the establishment of a protective and supportive legislative framework before the relevant technologies are developed, while the United States and China are focusing on research and intensive and even aggressive development at the expense of a legislative framework;

Amendment 37 #

Draft opinion
Recital E

E. Whereas ~~current policy and eth~~many ethical challenges related to AI applications and indicated in the Commission White Paper on Artificial ~~guidelines for AI~~Intelligence, are already addressed by the Union acquis, e.g. risk-assessment processes in place for AI-based health solutions in the Single Market; whereas other areas are lagging behind ethical challenges that must be identified and mitigated, since AI has tremendous capability to threaten patient preference, safety, and privacy; whereas the boundaries between the roles of physicians and machines in patient care need to be outlined;

Amendment 38 #

Draft opinion
Recital E

E. Whereas current policy and ethical guidelines for AI are lagging behind ethical challenges that must be identified and mitigated, since AI has tremendous capability to threaten patient preference, safety, and privacy; whereas the boundaries between the roles of physicians and machines in patient care need to be outlined in accordance with the principle of supervised robot autonomy, under which the initial programming of care and the final decision as to how it is to provide it, including diagnostics, remain within the decision-taking sphere of the appropriate physician or care personnel;

Amendment 39 #

Draft opinion
Recital E

Amendment 4 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Recital A b (new)

Ab. Whereas environment, health and food safety research and innovation activities carried out in Europe must comply with ethical principles and relevant national, Union and international legislation, including the Charter of Fundamental Rights of the European Union and the European Convention on Human Rights and its Supplementary Protocols1a; _________________ 1aArticle 19 of Regulation (EU) No 1291/2013 of the European Parliament and of the Council of 11 December 2013 establishing Horizon 2020 - the Framework Programme for Research and Innovation (2014-2020) and repealing Decision No 1982/2006/EC Text with EEA relevance

Amendment 40 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Recital E

Amendment 41 #

Róża THUN UND HOHENSTEIN

Draft opinion
Recital E

E. Whereas the current ~~policy~~legal framework of the Union and ethical guidelines for AI are lagging behind ethical challenges that must be identified and mitigated, since AI has tremendous capability to threaten patient preference, safety, and privacy; whereas the boundaries between the roles of physicians and machines in patient care need to be outlined;

Amendment 42 #

Draft opinion
Recital E a (new)

Ea. Whereas Union data protection rules should be adapted to take into account the increasing complexity and interconnectivity of care and medical robots that may handle highly sensitive personal information and health data and should be consistent with privacy by design as established by Regulation (EU) 2016/679 on data protection;

Amendment 43 #

Draft opinion
Recital E a (new)

Ea. Whereas solutions which stress the need to include scientific research as the basis for development strategies by creating repositories of medical data (e.g. neurological and cardiological data) and sharing data from this research can produce tangible social benefits in the context of public safety and health;

Amendment 44 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Susana SOLÍS PÉREZ

Draft opinion
Recital F

F. Whereas ~~AI solutions may benefit society in the area of food safety and Farming 2.0,~~data aggregation in connection with digital farming systems is handled by means of data platforms held by large business groups; whereas, accordingly, the digital tools used within farming tend to focus on mass production and disregard small and medium-sized producers; whereas the ~~Union holds leadership in AI applications~~high-tech solutions encouraged within farming encourage more intensive and standardised production, thus strengthening the global biodiversity trend;

Amendment 45 #

Draft opinion
Recital F

F. Whereas AI solutions may benefit society in the area of food safety and Farming 2.0~~, where the Union holds leadership in AI application~~ as it can directly reduce the use of pesticides, increase efficiency of farming, be used in applications such as automated machine adjustments for weather forecasting and disease identification, aid the challenges farmers face such as climate variation, and overall improve crop management practices;

Amendment 46 #

Draft opinion
Recital F

F. Whereas AI solutions may benefit society in the area of food safety, andmong others supporting precision farming or more broadly Farming 2.0, where the Union holds leadership in AI applications, and which will allow to combine more effective production with higher environmental standards and better utilization of resources;

Amendment 47 #

Draft opinion
Recital F

F. Whereas AI solutions may benefit society in the area of food safety and Farming 2.0, where the Union holds leadership in AI applications; whereas more generally AI solutions embedded in Digital Twins of the Earth may release the potential of data, where the Union holds a leadership with the Copernicus programme, in support of the Green Deal priority actions on climate change, circular economy, zero-pollution, biodiversity, deforestation and compliance assurance;

Amendment 48 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ

Draft opinion
Recital F

F. Whereas AI solutions may benefit society in the area of food safety and Farming 2.0, where the Union holds leadership in AI applications, especially in areas where water resources are scarce and climate change has severe impacts;

Amendment 49 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Susana SOLÍS PÉREZ, Martin HOJSÍK

Draft opinion
Recital F a (new)

Fa. Whereas, both globally and at European level, AI is one of the 21st century's strategic technologies; whereas in January 2017 the Chinese electrical- appliance business Midea completed acquisition of the European firm KUKA AG, one of the world's leading robotics, installation technology and intelligent- systems engineering suppliers; whereas that takeover should have been a wake-up call with regard to protecting Europe's AI sector, in particular SMEs and start-ups; whereas EU Competition Commissioner Margrethe Vestager has alerted Member States to the risks surrounding Chinese corporate acquisitions during the coronavirus crisis; whereas the Commission has pointed out that, in the context of the COVID-19 crisis, there could be a greater risk of attempts to acquire healthcare capacity (e.g. to produce medical devices or protective equipment) or facilities carrying out related activities such as research institutes (e.g. to develop vaccines) through foreign direct investment;

Amendment 5 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB

Draft opinion
Recital A c (new)

Ac. Whereas Article 16 TFEU states that everyone has the right to the protection of their personal data; whereas Article 22 of the General Data Protection Regulation refers to the situation where data are only used by automated processing, and recognises the right of the data subject not to be subject to a decision based solely on automated processing;

Amendment 50 #

Draft opinion
Recital F a (new)

Fa. Whereas the scope of that framework should be adequate, proportionate and thoroughly assessed; whereas it should cover a wide range of technologies and their components, including algorithms, software and data used or produced by AI; whereas a targeted approach based on the concept of high risk is necessary to avoid hampering future innovations in delivering benefits of AI applications e.g. in healthcare, environment protection and food quality to the citizens;

Amendment 51 #

Draft opinion
Recital F a (new)

Fa. Whereas it is essential to identify an effective means of ensuring trustworthy digital technologies, making it possible to reap their benefits while protecting fundamental rights and encouraging the development of informal, open, tolerant and just societies; this is particularly important in the case of hybrid human/artificial intelligence systems;

Amendment 52 #

Draft opinion
Recital F a (new)

Fa. Whereas there are more and more AI-technology-based e-health (mobile health) applications; whereas this area is growing exponentially (up 330% between 2014 and 2017) and represents a market that will have an estimated value of USD 223.7 billion in 2023;

Amendment 53 #

Draft opinion
Recital F b (new)

Fb. Whereas exponential growth in the development of AI and its incorporation into applications, products and services has been conducive to the emergence, in particular via the platform economy, of new business and service models; whereas work will be reorganised as a result of AI, producing a state of hyperconnectivity accompanied by more intensive work, increased stress and working time and a blurring of work-life boundaries; whereas this development has major consequences for workers' mental and physical health because it is a potential factor triggering occupational illnesses such as burnout or particular forms of social anxiety;

Amendment 54 #

Draft opinion
Recital F c (new)

Fc. Whereas the development of AI is conducive to the emergence of microtaskers working on on-line platforms whose tasks are digitally apportioned among them by algorithms; whereas platform microtasking is growing exponentially; whereas in 2013, according to the World Bank, there were 145 on-line work platforms, attracting 50 million crowdworkers; whereas those figures are now likely to be out of date; whereas platform microtasking is mainly carried out by vulnerable and precarious groups such as retired persons, students, unemployed people, migrants and people with disabilities; whereas these workers are constantly confronted with algorithmic reputation scores and automated tests and are constantly at risk of being disconnected from platforms, without redress, in the event of poor scores over an extended period; whereas they are therefore particularly exposed to precariousness, social isolation, burnout, and work and a pace of life that have no structure, with major consequences for their physical and mental well-being;

Amendment 55 #

Draft opinion
Recital F d (new)

Fd. Whereas robotic machines blur the boundaries between human subjects and technological objects; whereas not only do they have implications for society that need to be ethically assessed, but they also challenge the very ethical frameworks on the basis of which they are to be assessed; whereas, as is pointed out in the report by the World Commission on the Ethics of Scientific Knowledge and Technology (COMEST), particular attention should be paid to the use of medical robots, nursing robots, care robots for the elderly and companion robots;

Amendment 56 #

Draft opinion
Recital F e (new)

Fe. Whereas the use of social robots and companion robots is spreading rapidly within healthcare and, in particular, within elderly care; whereas care robots for the elderly and companion robots may take on a functional and emotional role; whereas those robots may have a role to play in reducing loneliness among older people, preventing behaviours associated with dementia, stimulating the cognitive activities of patients with a neurodegenerative disease or performing particular everyday tasks that are difficult for elderly persons to carry out; whereas companion robots may thus provoke feelings that are false, illusory and unreciprocated, deluding and infantilising older people;

Amendment 57 #

Draft opinion
Recital F f (new)

Ff. Whereas companion robots may increasingly be used for sexual purposes; whereas the use of sex robots that look like children or are programmed to be abused has particularly worrying ethical implications;

Amendment 58 #

Draft opinion
Paragraph 1

1. Stresses that the Union ~~must~~should establish a pan-European network of health experts and leading ethicists to work in a transparent manner to undertake all necessary steps to guarantee that its ethical values, as expressed in the acquis, apply effectively to all AI areas within its territory and to promote its standards worldwide;

Amendment 59 #

Draft opinion
Paragraph 1

1. Stresses that the Union must undertake all necessary steps to guarantee that its ethical values, as expressed in the acquis, apply effectively to all AI areas within its territory and to promote its standards worldwide; emphasises in this regard that technological developments in AI must always be to the benefit of humankind;

Amendment 6 #

Draft opinion
Recital B

B. Whereas the artificial intelligence (AI) global leadership race, which will ~~determine~~affect the source of ethical values and standards shaping the sector worldwide, is picking up the pace; whereas the integration of big data and AI technologies into public health systems and society must be accompanied by appropriate rules, standards and legislation that protect the fundamental rights of individuals and address new ethical challenges;

Amendment 60 #

Draft opinion
Paragraph 1 a (new)

1a. Considers it essential that AI, robotics and related technologies support the achievement of sustainable development, climate neutrality and circular economy goals; considers also that the use of these technologies should be environmentally friendly and support the achievement of environmental and public health goals, such as waste reduction, lowering the carbon footprint, reversing climate change and environmental degradation;

Amendment 61 #

Draft opinion
Paragraph 1 a (new)

1a. Stresses that Europe’s AI’s strategy should be consistent with SDGs and the objective of the 7th EAP to “live well within planetary boundaries”; stresses also that the environmental footprint of the deployment of AI, and its direct and indirect positive and negative impacts on energy and material consumption in Europe should be assessed vis-à-vis Europe’s commitment to the Paris Agreement, to carbon neutrality by 2050 and to a circular economy which aims to reducing Europe’s current material footprint;

Amendment 62 #

Draft opinion
Paragraph 1 a (new)

1a. Takes note of the first definition of AI proposed by the Commission in its April 2018 communication on artificial intelligence for Europe; notes the need to lay down a common definition of the concepts of robot and artificial intelligence that is technology-neutral and sufficiently flexible in order to ensure the legal certainty conducive to stimulating essential investment in this area in the EU and to ensuring public trust and consumer protection;

Amendment 63 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 1 a (new)

1a. Underlines the importance of the “right to explanation” of any decision taken by automated processing, this is the right to obtain human intervention on the part of the controller, to express his or her point of view and to contest the decision taken by an automated system4a; states, therefore, that ultimately, humans should keep the responsibility for decision making, especially in sectors where there are high stakes and risks such as health; _________________ 4a Article 22 GDPR Regulation

Amendment 64 #

Draft opinion
Paragraph 1 a (new)

1a. Underlines that the Union must undertake all necessary steps to increase the trust of the society in the development and implementation of AI, robotics and related technologies; in light of the significant impact that these technologies can have on citizens; calls on the Commission to follow the ethics guidelines on trustworthy AI and propose adequate measures to make sure that those technologies do not generate unfairly biased outputs for citizens;

Amendment 65 #

Draft opinion
Paragraph 1 a (new)

1a. Stresses the need for a regulatory framework stipulating the ethical principles to be applied to the design, development, implementation and functioning of this technology - from data access to strict outcome monitoring;

Amendment 66 #

Irena JOVEVA, Véronique TRILLET-LENOIR, Frédérique RIES, Martin HOJSÍK

Draft opinion
Paragraph 1 a (new)

1a. Points out, as stated by the Commission's Joint Research Centre, that AI remains a developing field which is not yet properly understood and that the very concept of AI has not yet been clearly defined;

Amendment 67 #

Amendment 68 #

Draft opinion
Paragraph 2

Amendment 69 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB

Draft opinion
Paragraph 2

2. Underlines that overregulation may hamper AI sector innovation, especially for SMEs and Start-ups; considers that hindering the Union AI sector in delivering benefits of AI applications in e.g. healthcare, environment protection and food quality to the citizens, may also bear ethical weight, especially in the context of global competition, where securing full respect of Union ethical values may pose a challengereminds that beyond Union action, many third countries are working on their ethical frameworks and that there are multiple proposals at a global level; is aware that the main difficulty for ethical principles may lie in the application of such principles rather than in their existence especially in the context of global competition and of high risk sectors as healthcare; recalls, in this regard, that Parliament’s resolution of 16 February 2017 on Civil Law Rules on Robotics asked the Commission to consider the designation of a European Agency for Artificial Intelligence that ensures a harmonised approach across the Union, develops common criteria and an application process relating to the granting of a European certificate of ethical compliance, and that addresses the new opportunities and challenges, in particular those of a cross-border nature, arising from ongoing technological developments;

Amendment 7 #

Draft opinion
Recital B

B. Whereas the artificial intelligence (AI) global leadership race~~, which~~ may well bring about a race to the bottom as regards national regulations; whereas, therefore, the European Union should grasp this opportunity to map out a regulatory framework that will determine the source of ethical values and standards shaping the sector worldwide~~, is picking up the pace~~;

Amendment 70 #

Draft opinion
Paragraph 2

2. Underlines th~~at overregulation may hamper~~e fact that legal safeguards may stimulate AI sector innovation, especially for SMEs and Sstart- ups; considers that hindering the Union AI sector in delivering benefits of AI applications in e.g. healthcare, environment protection and food quality to the citizens, may also bear ethical weight, especially in the context of global competition, where securing full respect of Union ethical values may pose a challengepoints out, especially in the context of global competition, that, with the GDPR, the European Union has led the way towards a binding ethically rigorous model;

Amendment 71 #

Draft opinion
Paragraph 2

2. Underlines th~~at overregulation may hamper AI sector innovation, especially for~~e fact that AI R&D is a fast-expanding sector that is attracting a very large number of SMEs and Sstart-ups~~; considers that hindering the Union AI sector in delivering~~, but also private and public investment; underscores the benefits of AI applications in ~~e.g.~~ healthcare, environment protection and food quality ~~to the citizens, may also bear ethical weight, especially in the context of global competition, where securing full respect of Union ethical values may pose a challeng~~, for example;

Amendment 72 #

Draft opinion
Paragraph 2

Amendment 73 #

Draft opinion
Paragraph 2 a (new)

2a. Considers that for all AI applications, developed in the Union and outside of it, the same level of protection must be secured in the Union as it is with all other technologies, including effective judicial redress for parties negatively affected by AI systems, whilst technological innovation needs to be allowed to continue to develop; considers furthermore that this AI risk area is crucial e.g. for the health services, transport involving autonomous vehicles and food safety; calls for a clear distribution of obligations, rights and liabilities among the economic operators involved in AI applications delivery, to attribute each obligation to the actor(s) who is (are) best placed to address any potential risks, whether this is the developer, the deployer, the producer, the distributor or importer, the service provider, the professional or private user, and in this regard for adequate revision of relevant EU legislation, e.g. of the Product Liability Directive and for the harmonization of national legislation; supports the Commission position expressed in the White Paper that due to the complexity of AI systems, securing effective level of protection and redress may require adapting the burden of proof required by national rules on liability for damage caused by the operation of AI applications; is of the opinion that clarity as to legal liability in the AI sector will strengthen enforcement of Union ethical values embodied in its acquis, legal certainty and predictability, and social acceptance supporting the development of a Union AI ecosystem of excellence by pooling investors and increasing market uptake;

Amendment 74 #

Draft opinion
Paragraph 2 a (new)

2a. Points out that the OECD's ethical framework takes account of labour market upheaval; stresses that automation combined with AI will increase productivity and therefore increase output; points out that, as during previous technological revolutions, some jobs will be replaced; stresses that increased use of robotics and AI should also reduce human exposure to harmful and hazardous conditions and should also help to create more quality and decent jobs and improve productivity; points to the work of the OECD, which stresses that automation may give society the option to cut the number of hours worked, thus improving workers' living conditions and health;

Amendment 75 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Paragraph 2 a (new)

2a. Highlights that many of the proposals by countries which are not members of the Union and by international organizations revolve around common principles or concepts for AI, those being: human-centricness, trustworthiness, respect for human autonomy, harm prevention, equity and "no one left behind" and explicability; is of the opinion that an international ethical framework around these principles would be highly desirable; is concerned about AI progress and innovations leading to social inequality if no action is taken; calls therefore on the Commission and Member States to take the necessary measures to leave no one behind in the transition to a digital Europe, and to guarantee a fair, affordable and equal access to these innovations especially in areas such as healthcare;

Amendment 76 #

Draft opinion
Paragraph 2 a (new)

2a. Notes that the law of Member States is not harmonised with regard to liability rules that are applicable to damage or injury that could result from the application of emerging digital and behavioural technologies; calls on the Commission and Member States to focus on the vulnerability and liability gaps in AI, robotics and related technologies as well as on adjustments that need to be made to existing Union legislation, such as the Product Liability and Medical Devices Directives;

Amendment 77 #

Draft opinion
Paragraph 2 a (new)

2a. Points out that the safety standards laid down in the Medical Devices Regulation may not be sufficient for the challenges of the mass employment of artificial intelligence systems; calls on the Commission to monitor the challenges in this field and to put forward proposals where necessary;

Amendment 78 #

Draft opinion
Paragraph 2 b (new)

2b. Stresses that social dialogue must be used to ensure a fair transition for workers as AI is deployed and tasks become automated, and this transition should include access to training programmes throughout their working lives, support for those affected by job cuts and access to new opportunities on the labour market; stresses that firms should also invest in training and reskilling their existing workforce with a view to addressing their needs;

Amendment 79 #

Draft opinion
Paragraph 2 b (new)

2b. Emphasises that AI applications in health should always have the aim of maximising the opportunities they can bring – such as improving the health of individual patients as well as the performance of Member States’ public health systems – without lowering ethical standards and without threatening the privacy or safety of citizens;

Amendment 8 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV, Tiemo WÖLKEN

Draft opinion
Recital B

B. Whereas the artificial intelligence (AI) global leadership race, which will determine the ~~source of~~ ethical values and standards shaping the sector worldwide, is picking up ~~the~~ pace and the Union should set an example for the rest of the world;

Amendment 80 #

Draft opinion
Paragraph 2 c (new)

2c. Draws further attention to the OECD recommendations calling for governments to work closely with stakeholders to promote the responsible use of AI at work, to enhance the safety of workers and the quality of jobs, and to aim to ensure that the benefits of AI are broadly and fairly shared;

Amendment 81 #

Draft opinion
Paragraph 3

3. Welcomes the fact that the Risk- Based Approach methodology defined in the Commission White Paper of 19 February 20201 recognises healthcare, transport and energy as high risk sectors by default, and proposes ~~to introduce listed AI requirements beyond existing-Union~~horizontal rules inand the~~se sectors, unless the manner in which AI is used does not involve significant risk; stresses that the Union AI ethical framework should address especially the above high-risk~~ introduction of a list of more stringent AI requirements beyond existing Union rules in these sectors; _________________ 1White Paper On Artificial Intelligence - A European approach to excellence and trust, COM(2020)0065

Amendment 82 #

Draft opinion
Paragraph 3

3. ~~Welcom~~Notes the fact that the Risk- Based Approach methodology defined in the Commission White Paper of 19 February 20201 recognises healthcare, transport and energy as high risk sectors by default, and proposes to introduce listed AI requirements beyond existing-Union rules in these sectors~~, unl~~; recalls the vagueness of the ~~manner in which AI is used does not involve significant risk; stresses that the Union AI ethical framework should address especially the above high-risk sectors~~sector-specific approach outlined in the White Paper; underlines that there are many kinds of risk and levels of risk that evolve over time through more advanced future technologies; stresses that the Union AI ethical framework should address all sectors that adopt the use of these technologies as having equivalent levels of risk, rather than the two-tier (high/low) risk approach proposed; _________________ 1White Paper On Artificial Intelligence - A European approach to excellence and trust, COM(2020)0065

Amendment 83 #

Draft opinion
Paragraph 3

3. Welcomes the fact that the Risk- Based Approach methodology defined in the Commission White Paper of 19 February 20201 recognises healthcare, transport and energy as high risk sectors by default, ~~and proposes to~~ introduceing listed AI requirements beyond existing-Union rules in these sectors, unless the manner in which AI is used does not involve significant risk; stresses that the Union AI ethical framework should address especially the above high-risk sectors; _________________ 1White Paper On Artificial Intelligence - A European approach to excellence and trust, COM(2020)0065

Amendment 84 #

Draft opinion
Paragraph 3 a (new)

3a. Stresses that the development and use of AI, including deep learning methods, has a significant environmental impact because it increases overall energy consumption, the volume of data stored, and the amount of electric and electronic waste; proposes that two measures be introduced in the new Circular Economy Action Plan – the European waste- reduction and resource-efficiency strategy: the setting of an overall target of halving the EU's material footprint by 2030 and of a ceiling for absolute reduction of waste per capita, by carrying out direct action to prevent waste generation and introducing a specific sectoral target for some waste, such as electronic waste;

Amendment 85 #

Draft opinion
Paragraph 3 a (new)

3a. Calls for clear, objective and transparent procedures at Union level for establishing a public catalogue of AI high-risk applications involving a periodic review and update mechanism; calls for consideration of putting the burden of proof in such procedures, for all AI applications in all domains, on the entity seeking to develop or deploy the AI system, in order to maintain the catalogue open for innovation and avoid ignoring the risk of classifying AI applications as being no high risk;

Amendment 86 #

Draft opinion
Paragraph 3 b (new)

3b. Recommends supplementing the Risk-Based Approach with an Algorithmic Impact Assessment drawing information for example from the Regulatory Impact Analysis (RIA), GDPR Risk Assessment Procedure, Human Rights Impact Assessment (HRIA) with the results made publicly viewable;

Amendment 87 #

Draft opinion
Paragraph 3 b (new)

3b. Highly recommends that the Member States and the Commission promote the use of due diligence in the trade of metals needed to build electronic and robotics technologies;

Amendment 88 #

Draft opinion
Paragraph 3 c (new)

3c. Insists that European green data centres should be promoted and developed with a view to reducing our dependency on foreign and private data centres, subject to European rules and standards, with a medium-term objective of 100% renewable energy use;

Amendment 89 #

Draft opinion
Paragraph 3 c (new)

3c. Welcomes the voluntary labelling initiative for non-high risk AI;

Amendment 9 #

Draft opinion
Recital B

B. Whereas the global competition in artificial intelligence (AI) ~~global leadership race~~development, which will determine the source of ethical values and standards shaping the sector worldwide, is picking up the pace;

Amendment 90 #

Draft opinion
Paragraph 3 d (new)

3d. Stresses the scale of the ecological footprint of AI and robotics, as discussed in guideline 6 of the Commission communication 'Building Trust in Human-Centric Artificial Intelligence'; insists that the environmental impact of these areas needs to be limited through energy-efficient consumption, by promoting the use of renewable technologies for robotics and the use and re-use of secondary raw materials, and by cutting down on electric and electronic waste; encourages the Commission, therefore, to incorporate circular economy principles into all the EU's robotics policies;

Amendment 91 #

Irena JOVEVA, Nils TORVALDS, Véronique TRILLET-LENOIR, Martin HOJSÍK

Draft opinion
Paragraph 3 d (new)

3d. Welcomes the Commission commitment expressed in the White Paper to examine safety and liability challenges that are distinctive to healthcare e.g. AI systems providing specialized medical information to physicians or directly to the patient, AI systems performing medical tasks themselves directly on a patient; calls for corresponding examination of the other listed sectors that are by default high-risk ones;

Amendment 92 #

Amendment 93 #

Draft opinion
Paragraph 4

4. Notes that, ~~with~~due to the fact that legal regulations respond better to current well-defined challenges and due to the rapid development of AI ~~and~~resulting in the uncertainty that lies ahead, a common, legally well-anchored, Union AI ethical framework ~~will~~may expand an ecosystem of trust as defined in the Commission White Paper, whether in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and predictability, providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories;

Amendment 94 #

Sara CERDAS

Draft opinion
Paragraph 4

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust as defined in the Commission White Paper, whether in environment ~~protection, healthcare~~or public health protection, the creation of healthier environments, better healthcare resources and services or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories;

Amendment 95 #

Draft opinion
Paragraph 4

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework ~~will expand~~that is enforceable and binding should develop an ecosystem of trust as defined in the Commission White Paper, whether in environmental protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and ethical excellence in providing an effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories;

Amendment 96 #

Draft opinion
Paragraph 4

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust as defined in the Commission White Paper, whether in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories; points out that the Copernicus programme can serve as a best practice in developing high quality large datasets as input to AI models;

Amendment 97 #

Draft opinion
Paragraph 4

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust, including for users, as defined in the Commission White Paper, ~~whether~~particularly in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories;

Amendment 98 #

César LUENA, Manuel PIZARRO, Javi LÓPEZ, Tudor CIUHODARU, Rovana PLUMB, Sara CERDAS, Milan BRGLEZ, Łukasz KOHUT, Romana JERKOVIĆ, Cristina MAESTRE, Petar VITANOV

Draft opinion
Paragraph 4

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust as defined in the Commission White Paper, whether in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories; points out that ethics is not made up of permanent principles, but has been changing over the course of various cultures and times; supports in this regard that the framework should be periodically reviewed to guarantee its applicability through time and new developments;

Amendment 99 #

4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust as defined in the Commission White Paper, whether in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories; therefore underlines the importance of a human- centric approach and of regular reviews on AI advances, by sector, in order to promote proactive regulation;

source: 652.646

2020/06/15 LIBE 116 amendments...

Amendment 1 #

Draft opinion
Citation -1 (new)

– having regard to Articles 2 and 3 of the Treaty on European Union (TEU),

Amendment 10 #

Roberta METSOLA, Juan Ignacio ZOIDO ÁLVAREZ, Javier ZARZALEJOS, Kris PEETERS, Axel VOSS

Draft opinion
Citation -1 i (new)

– having regard to the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions of 11 December 2019 on The European Green Deal,

Amendment 100 #

Draft opinion
Paragraph 8

8. Stresses that AI and robotics are not immune from making mistakes; emphasises the importance of the right to an explanation when persons are subjected to algorithmic decision-making; considers the need for legislators to reflect upon the complex issue of liability in the context of criminal justice.

Amendment 101 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 8

8. Stresses that AI and robotics are not immune from making mistakes; considers the need for legislators to reflect upon the complex issue of liability in the context of both civil and criminal justice.

Amendment 102 #

Birgit SIPPEL, Katarina BARLEY

Draft opinion
Paragraph 8

8. ~~Stresses that AI and robotics are not immune from making mistakes; c~~Considers the need for legislators to reflect upon the complex issue of liability in the context of criminal justice. and that liability in all AI applications should always rest with a legal person;

Amendment 103 #

Kris PEETERS, Axel VOSS

Draft opinion
Paragraph 8

8. Stresses that AI and robotics are not immune from making mistakes; considers the need for legislators to reflect upon the complex issue of liability ~~in the context of criminal justice~~.

Amendment 104 #

Draft opinion
Paragraph 8 a (new)

8a. Reminds that according to Article 22 of the General Data Protection Regulation, a person has the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning him or her or similarly significantly affects him or her and that only very limited exceptions exist to this general rule; recalls the need for algorithms to be transparent, especially those that produce legal effects concerning individuals; underlines that transparency about the underlying logic of an algorithm is highly relevant for the affected individual in order for his or her fundamental rights to be fully protected; reminds, in addition, that transparency about the algorithms used is also of utmost relevance for the person taking a final decision based on an algorithmic calculation, for example, when a bank employee has to decide on a loan application or a human resources professional on whom to hire.

Amendment 105 #

Draft opinion
Paragraph 8 a (new)

8a. Proposes to establish a European Agency for Artificial Intelligence to coordinate the determination of breaches of the principles of precautionary and of non-discrimination by competent supervisory authorities, in cooperation with other competent authorities in the Union, notably the Consumer Protection Cooperation Network, national consumer protection bodies, the high level group on Non-discrimination, Equality and Diversity, and civil society, and facilitate means for individuals to meaningfully contest and remedy harm caused by such discrimination, and other infringement of fundamental rights by artificial intelligence and automated decision making systems, whether stemming from public or private sector actors;

Amendment 106 #

Draft opinion
Paragraph 8 a (new)

8a. Notes that AI is often used by digital platforms for automated decision making algorithms to disseminate and order the content shown to the users, including to organise their personal feed; stresses that these algorithms, how they work and how they order the shown material, are a black box to users, which takes away choice and control from the user, enables the creation of echo chambers and leads to a distrust in digital services; calls on the Commission to take the perspective of the end-user of AI- applications into account while drafting its ethical framework, with a focus on transparency, explicability and customization.

Amendment 107 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 8 a (new)

8a. Underlines that artificial intelligence, robotics and related technologies are global technologies and that these standards need to be adopted worldwide in order to ensure their future development is aligned to European values and ethical standards; calls on the Commission to engage in AI diplomacy in international fora with likeminded partners such as the United States, the G7, the G20, and OECD for establishing common ethical standards and guidelines for developing AI, robotics, and related technologies;

Amendment 108 #

Draft opinion
Paragraph 8 a (new)

8a. Reiterates the call for the establishment of a European Agency for Artificial Intelligence, and emphasises the importance of having national supervisory authorities in each Member State responsible for ensuring, assessing and monitoring compliance with ethical principles and legal obligations pertaining to the development, deployment and use of artificial intelligence, robotics and related technologies.

Amendment 109 #

Draft opinion
Paragraph 8 a (new)

8a. Stresses that a clear framework needs to be introduced for the use of AI by social media platforms, as do transparency requirements for the algorithms used and the calibration thereof, in order to prevent excessive content-removal and any form of filtering or censorship of the internet;

Amendment 11 #

Draft opinion
Citation -1 j (new)

– having regard to its resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics1a, __________________ 1a OJ C 252, 18.7.2018, p. 239.

Amendment 110 #

Draft opinion
Paragraph 8 b (new)

8b. Notes that AI can be used to analyse or predict aspects concerning individual's personal preferences, interests or behaviour, for profiling; emphasises that the quality of output of automated decision making AI is subject to the quality of used data and the chosen predetermined parameters; stresses that the use of automated decision making AI requires a strong legislative framework which protects privacy and personal data, and together with a duty of care obligation overseeing the legitimate use of the AI, which does not apply to content moderation, ensures full compliance; calls therefore on the Commission to work out a duty of care regime through detailed sectoral guidelines in order to use automated decision making algorithms in compliance with the fundamental rights of protection of personal data and privacy, laid down in the General Data Protection Regulation1a; __________________ 1aRecital 71 and article 22 General Data Protection Regulation.

Amendment 111 #

Draft opinion
Paragraph 8 b (new)

8b. Stresses that the deployment of AI, robotics, and related technologies in society relies on digital infrastructure that needs to be highly secure, resilient, and free from potential tampering; calls on the Commission to ensure the security and resilience of the European digital infrastructure by reducing reliance on technology produced by companies originating in countries and influenced by regimes that do not share our values and respect for human rights;

Amendment 112 #

Draft opinion
Paragraph 8 b (new)

8b. Suggests to create a centre of expertise, bringing together academia, research, industry, and individual experts at Union level, either as an integral part of or associated with such Agency, to foster exchange of knowledge and technical expertise, and to facilitate collaboration throughout the EU and beyond;

Amendment 113 #

Draft opinion
Paragraph 8 c (new)

8c. Points out that AI can be used for personalised advertising, in particular micro-targeted and behavioural advertisements, and of assessments of individuals; notes the potential negative impact of personalised advertisement, especially on minors, by interfering in the private life of individuals; posing questions as to the collection and use of the data used to personalise advertising, offering products or services or setting prices; perceives these downsides as expressions of the way personalised ads are unethical; calls therefore on the Commission to use its ethical framework to prohibit all personalised advertisements;

Amendment 114 #

Draft opinion
Paragraph 8 d (new)

8d. Notes that AI can be used to manipulate face- and audiovisual characteristics, often referred to as deepfakes; recalls that this technique can be used to manipulate elections, to disseminate disinformation and for other undesirable actions; asks the Commission therefore to use its ethical framework to impose an obligation for all deepfake material or any other realistically made synthetic videos, to state it's not original and to introduce a strict limitation when used for electoral purposes;

Amendment 115 #

Draft opinion
Paragraph 8 e (new)

8e. Emphasises the importance of the protection of personal data and privacy; observes the rapid development of AI applications to recognise unique characteristic elements, such as facial, movements and attitudes; warns for interferences of privacy, non- discrimination and the protection of personal data with the use of automated recognition applications; calls on the Commission to incorporate in its ethical framework an absolute ban on facial recognition in the public space and educational premises and a ban on not local storage of data used for facial recognition;

Amendment 116 #

Draft opinion
Paragraph 8 f (new)

8f. Recalls the importance of linguistic and cultural diversity; calls therefore on the Commission to use its ethical framework to not let AI reduce this diversity, but to keep offer access to a wide variety of content which would not over- represent a single language and/or cultural model and to condemn any attempts from algorithms which would restrict this diversity and only offer content corresponding to some already existing patterns or which could act as an 'echo-chamber' that would prevent access to more diversity;

Amendment 12 #

Draft opinion
Citation -1 k (new)

– having regard to the OECD Council Recommendation on Artificial Intelligence adopted on 22 May 2019,

Amendment 13 #

Draft opinion
Recital A (new)

A. whereas the development and design of so-called ‘artificial intelligence’, robotics and related technologies is done by humans, and their choices determine the potential of technology to benefit society;

Amendment 14 #

Draft opinion
Recital B (new)

B. whereas algorithmic accountability should mean implementing technical and operational measures that ensure transparency, clearly assigned chains of responsibility, non- discrimination through automated decision-making or through calculating of probabilities of individual behaviour; whereas transparency should give individuals meaningful information about the logic involved, the significance and the envisaged consequences; whereas this should include information about the data used for training AI and allow individuals to understand and monitor the decisions affecting them;

Amendment 15 #

Draft opinion
Recital C (new)

C. whereas there are serious concerns that the current EU legal framework, including the consumer law acquis, product safety and market surveillance legislation, as well as antidiscrimination legislation is not always fit for purpose to effectively tackle the risks created by artificial intelligence, robotics and related technologies;

Amendment 16 #

Draft opinion
Recital D (new)

D. whereas artificial intelligence, robotics and related technologies can have serious implications for material and immaterial goods of individuals, groups, and society as a whole, and these individual and collective harms must be reflected in legislative responses;

Amendment 17 #

Draft opinion
Recital E (new)

E. whereas governance issues with the deployment of AI in the public sector must be duly considered in terms of its implications for democracy, especially democratic legitimacy, accountability, meaningful public engagement and oversight;

Amendment 18 #

Draft opinion
Recital F (new)

F. whereas data analysis and AI increasingly impact on the information made accessible to citizens; whereas such technologies, if misused, may endanger fundamental rights to information as well as media freedom and pluralism;

Amendment 19 #

Draft opinion
Recital G (new)

G. whereas ethical guidance, such as the principles adopted by the High-Level Expert Group on Artificial Intelligence, provides a good starting point but is not enough to ensure that businesses act fairly and guarantee the effective protection of individuals;

Amendment 2 #

Draft opinion
Citation -1 a (new)

– having regard to Articles 10, 19, 21 and 167 of the Treaty on the Functioning of the European Union (TFEU),

Amendment 20 #

Draft opinion
Paragraph -1 a (new)

-1a. Stresses that the prospects and opportunities of artificial intelligence can only be fully tapped into by citizens, the public and private sectors, academia and the scientific community when public trust in these technologies is ensured by a strong enforcement of fundamental rights and compliance with current EU data protection law and legal certainty for all actors involved; stresses that the processing of personal data can only be done pursuant to any of the legal bases laid down in Article 6 of Regulation (EU) 2016/679; considers that it is crucial that transparency and the proper provision of information to the audiences concerned are key to building public trust and to the protection of individual rights;

Amendment 21 #

Draft opinion
Paragraph -1 b (new)

-1b. Underlines that compliance with the existing data protection legislation, together with strong scientific, ethical and legal standards, and methods for democratic oversight, are key to establishing trust in and the reliability of AI solutions; further emphasises that information revealed by AI does not offer an impartial overview of any subject matter and is only as reliable as the underlying data permits; highlights that predictive analysis based on AI can only offer a statistical probability and therefore cannot always accurately predict individual behaviour; stresses, therefore, that strong scientific, ethical and legal standards are vital for managing data collection and judging the results of such AI analysis;

Amendment 22 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 1

1. Believes that ~~any ethical framework should seek to respect human autonomy, prevent harm, promote fairness, and respect the principle of explicability of technologies;~~ the development, deployment and use of Artificial Intelligence (AI), and thus any ethical framework, need to fully respect fundamental rights, freedoms and values, including privacy, the protection of personal data, non-discrimination and the freedom of expression and information, as enshrined in the Union data protection law, namely Regulation (EU) 2016/679 of the European Parliament and of the Council (GDPR)1a and Directive (EC) 2002/58 of the European Parliament and of the Council (ePrivacy)1b, currently under revision, and the Charter of Fundamental Rights of the European Union; calls on the Commission to incorporate an obligation of transparency, explicability and the possibility of human intervention of AI applications in its ethical framework and to take care in its legislative proposals of sufficient oversight, enforcement with effective penalties, independent audits, specific stress tests and all other necessary means for a proper functioning checks and balances system; __________________ 1aRegulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1). 1bDirective 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications) (OJ L 201, 31.7.2002, p. 37).

Amendment 23 #

Draft opinion
Paragraph 1

1. Believes that any ethical framework should ~~seek to~~ensure respect for human autonomy, ensure benefits for all, prevent harm, and promote fairness, ~~and respect the principle of explicability of technologies;~~ equality and transparency; notes that the potential for artificial intelligence (AI), robotics and related technologies that are truly ethical will inevitably conflict with the profit- orientation of private companies and interests; stresses therefore that an ethical framework for AI, robotics and related technologies is no substitute for wide- ranging and binding legal regulation of same; calls for the project of full and binding legal regulation of AI, robotics and related technologies by the European Union to be moved forward without any delay;

Amendment 24 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Birgit SIPPEL, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Elena YONCHEVA

Draft opinion
Paragraph 1

1. Believes that ~~any ethical framework should~~there is a difference between ethics and law and the role they play in our societies; any framework of ethical principles for the development, deployment and use of Artificial Intelligence (AI), robotics and related technologies should complement the EU Charter of Fundamental Rights and thereby seek to respect human dignity and autonomy, prevent harm, promote fairness, and transparency, respect the principle of explicability of technologies; and guarantee that the technologies are there to serve people, with the ultimate aim of increasing human well-being for everybody;

Amendment 25 #

Draft opinion
Paragraph 1

1. Believes that any ethical and legal framework should ~~seek to respect human autonomy~~respect and preserve human dignity, autonomy, and self- determination of the individual, prevent harm, promote fairness, and ~~respec~~implement the principles of explicability, interpretability, auditability, as well as traceability, transparency, and accessibility of technologies;

Amendment 26 #

Draft opinion
Paragraph 1

1. Believes that any ethical framework should seek to respect human autonomy, prevent harm, promote fairness and inclusion - especially of citizens living with disabilities -, fight discrimination, also of minority groups, and respect the principle of explicability of technologies;

Amendment 27 #

Draft opinion
Paragraph 1

1. Believes that any ethical framework should ~~seek to~~fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties; respect human autonomy,; prevent harm,; promote fairness, eliminate biases and discrimination; and respect the principle of explicability ~~of technologies~~;

Amendment 28 #

Draft opinion
Paragraph 1

1. Believes that any ethical framework should seek to respect human autonomy, prevent harm, promote fairness, and respect the principle of explicability of technologies; believes that technology should serve humans, and not replace or decide for them;

Amendment 29 #

Andor DELI

Draft opinion
Paragraph 1

1. Believes that any ethical framework should seek to respect human autonomy, prevent harm, promote fairness, and respect the principle of technological neutrality and explicability of technologies;

Amendment 3 #

Draft opinion
Citation -1 b (new)

– having regard to the right to petition enshrined in Articles 20 and 227 of the TFEU and Article 44 of the Charter of Fundamental Rights of the European Union (EUCFR),

Amendment 30 #

Draft opinion
Paragraph 1 – indent 1 (new)

- Independent audits on the application of AI should be conducted annually, in analogy with the financial sector, to examine whether the used AI- applications and checks and balances are in accordance with specified criteria; the auditors need to be supervised by an independent sufficiently overseeing authority;

Amendment 31 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA

Draft opinion
Paragraph 1 a (new)

1a. Warns that, owing to the intrusiveness of the decisions and measures taken by law enforcement authorities – including by means of data processing and AI – into the lives and rights of citizens, maximum caution is required in order to prevent unlawful discrimination and the targeting of certain individuals or groups of people defined by reference to race, colour, ethnic or social origin, genetic features, language, religion or belief, political or any other opinion, property, birth, disability, age, gender, gender expression or identity, sexual orientation, residence status, health or membership of a national minority which is often the subject of ethnic profiling or more intense law enforcement policing, as well as individuals who happen to be defined by particular characteristics; calls for proper training for the frontline collectors of data and users of intelligence derived from AI;

Amendment 32 #

Draft opinion
Paragraph 1 a (new)

1a. Emphasises that the development and deployment of AI, robotics and related technologies should not be left solely or principally to the private sector; stresses the urgent need to mobilise both the Union’s resources and the resources of Member States to work toward the creation of truly public, non-proprietary and ethical AI, robotics and related technologies, bearing in mind that AI in particular is a general purpose technology which is currently underpinning and will increasingly underpin critical public and social infrastructure in the future;

Amendment 33 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 1 a (new)

1a. Recalls that AI may give rise to biases and thus to various forms of discrimination, such as sex, race, colour, ethnic or social origin, genetic features, language, religion or belief, political or any other opinion, membership of a national minority, property, birth, disability, age or sexual orientation; recalls that everyone’s rights must be ensured and that this principle of non- discrimination must be the core of the ethical framework of the Commission;

Amendment 34 #

Draft opinion
Paragraph 1 b (new)

1b. Stresses that the development of AI, robotics and related technologies poses risks for human rights - namely privacy, data protection, and freedom of expression and information - and that in the future it may pose further risks that are still unknown; calls for the precautionary principle to be at the heart of both ethical and legal frameworks for AI;

Amendment 35 #

Clare DALY, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA

Draft opinion
Paragraph 2

~~2. Stresses the importance of developing an “ethics-by-default and by design” framework which fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties;~~deleted

Amendment 36 #

Draft opinion
Paragraph 2

2. Stresses the importance of developing an “ethics-by-default and by design” framework which fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties; stresses that European standards for AI must be based on the principles of digital ethics, human dignity, respect for fundamental rights, data protection and security; emphasises the importance of capitalising on the EU’s potential for creating a strong infrastructure for AI systems rooted in high standards of data and respect for humans; stresses that specific risk assessments, rather than broad sector-oriented criteria, should determine the level of risk of any AI system; calls for the introduction of a mandatory human rights impact assessment in the design and ongoing development of every AI system, including an evaluation of the societal implications of and risks posed by the system and an outline of the actions needed to mitigate such risks;

Amendment 37 #

Kris PEETERS, Juan Ignacio ZOIDO ÁLVAREZ, Roberta METSOLA, Axel VOSS

Draft opinion
Paragraph 2

2. Stresses the importance of developing an “ethics-by-default and by design” framework which fully respects the Charter of Fundamental Rights of the European Union, Union law and the Treaties; calls, in this regard, for a clear and coherent governance model that allows companies to further develop artificial intelligence, robotics and related technologies;

Amendment 38 #

Draft opinion
Paragraph 2

2. Stresses the importance of developing an “ethics-by-default and by design” legal framework which ensures that any AI put into operation fully respects the Charter of Fundamental Rights of the European Union, Union law and the Treaties; considers that this is in line with the precautionary principle that guides EU legislation;

Amendment 39 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Birgit SIPPEL, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Elena YONCHEVA

Draft opinion
Paragraph 2

2. SHighlights the power asymmetry between those who employ AI technologies and those who interact and are subject to them; in this context stresses the importance of developing an “ethics-by-default and by design” framework which fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties;

Amendment 4 #

Draft opinion
Citation -1 c (new)

– having regard to Articles 21 and 22 of the EUCFR,

Amendment 40 #

Axel VOSS

Draft opinion
Paragraph 2

Amendment 41 #

Draft opinion
Paragraph 2

2. Stresses th~~e importance of developing~~at citizens' trust in AI can only be built on an “ethic~~s-by-default and by design”~~al framework which fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties;

Amendment 42 #

Draft opinion
Paragraph 2 a (new)

2a. Calls on the European Union and on the Member States to promote public awareness of the risks and opportunities of the use of AI as an ethical requirement.

Amendment 43 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 3

3. Considers that the Union legal framework ~~will need to be updated with guiding ethical principles; points out that, where it would be premature to adopt legal acts, a soft law framework should be used~~, including the consumer law acquis, product safety and market surveillance legislation, as well as antidiscrimination legislation, will need to be updated to effectively tackle the risks created by artificial intelligence, robotics and related technologies; considers that ethical guidance constitutes a good starting point but must be reflected in binding legislative instruments in order to ensure that businesses act fairly and to effectively guarantee protection of individuals;

Amendment 44 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Birgit SIPPEL, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Elena YONCHEVA

Draft opinion
Paragraph 3

3. Considers that the current Union legalislative framework ~~will need to be upda~~on protection of privacy and personal data fully applies to AI, robotics and related technologies, however could benefit from being supplemented with ~~guiding~~robust ethical ~~principl~~guidelines; points out that, where it would be premature to adopt legal acts, a soft law framework should be used;

Amendment 45 #

Draft opinion
Paragraph 3

3. Considers that the Union legal framework ~~will~~may need to be ~~upda~~complemented with guiding ethical principles; ~~points out that, where it would be premature to adopt legal acts, a soft law framework should be used;~~

Amendment 46 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 3

3. Considers that the Union legal framework will need to be flexible, future- proof, and continuously updated with guiding ethical principles; points out that, where it would be premature to adopt legal acts, a soft law framework should be used;

Amendment 47 #

Andor DELI

Draft opinion
Paragraph 3

3. Considers that the Union legal framework will need to be revised and updated with guiding ethical principles; points out that, where it would be premature to adopt legal acts, a soft law framework should be used;

Amendment 48 #

Draft opinion
Paragraph 3 a (new)

3a. Recalls that the lack of transparency of AI systems makes it difficult to identify and prove possible breaches of laws, including legal provisions that protect fundamental rights; believes that an examination of, and guidelines on, how the Union’s human rights frameworks and the obligations that flow therefrom can protect citizens in the context of the widespread use of AI, robotics and related technologies are urgently needed; stresses the need to assess whether the EU’s human rights framework will need to be updated to meet the challenge posed to rights by these complex and emergent technologies;

Amendment 49 #

Roberta METSOLA, Javier ZARZALEJOS, Kris PEETERS

Draft opinion
Paragraph 3 a (new)

3a. Expects the Commission to integrate a strong ethical framework into the forthcoming legislative proposal as a follow up to the White Paper on Artificial Intelligence, including on safety, liability, fundamental rights and data protection, which maximises the opportunities and minimises the risks of AI technologies;

Amendment 5 #

Kris PEETERS, Juan Ignacio ZOIDO ÁLVAREZ, Roberta METSOLA, Axel VOSS

Draft opinion
Citation -1 d (new)

– having regard to the preamble to the TEU,

Amendment 50 #

Draft opinion
Paragraph 3 a (new)

3a. Calls on the commission to consider developing a framework of criteria and indicators to label AI technology, in which developers could participate voluntarily, in order to stimulate comprehensibility, transparency, accountability and incentivise additional precautions by developers;

Amendment 51 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 3 b (new)

3b. Stresses the need to assess how existing EU rules, in particular data protection rules, apply to AI and how proper enforcement of these rules in this field can be assured; calls on the Commission, the Member States and the data protection authorities to identify and take any possible measures to minimise algorithmic discrimination and bias and to develop a strong and common ethical framework for the transparent processing of personal data and automated decision- making that can guide data usage and the ongoing enforcement of Union law;

Amendment 52 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 3 b (new)

3b. Expects that the forthcoming legislative proposal will include policy solutions to the major recognised risks of Artificial Intelligence including, amongst others, on the ethical collection and use of Big Data, the issue of algorithmic transparency and algorithmic bias;

Amendment 53 #

Draft opinion
Paragraph 3 c (new)

3c. Stresses that the data sets and algorithmic systems used when making classifications, assessments and predictions at the different stages of data processing in the development of AI, robotics and related technologies may result not only in infringements of the fundamental rights of individuals, but also in differential treatment of and indirect discrimination against groups of people with similar characteristics; calls for a rigorous examination of AI’s politics and consequences, including close attention to AI’s classification practices and harms; emphasises that ethical AI, robotics and related technologies require that the field centre non-technical disciplines whose work traditionally examines such issues, including science and technology studies, critical race studies, disability studies, and other disciplines attuned to social context, including how difference is constructed, the work of classification, and its consequences; stresses the need therefore to systematically and immediately invest in integrating these disciplines into AI study and research at all levels;

Amendment 54 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 3 d (new)

3d. Notes that the field of AI, robotics and related technologies is strikingly homogenous and lacking in diversity; recognises the need to ensure that the teams that design, develop, test, maintain, deploy and procure these systems reflect the diversity of its uses and of society in general in order to ensure that bias is not unwittingly ‘built in’ to these technologies;

Amendment 55 #

Draft opinion
Paragraph 4

4. Is of the opinion that effective cross- border cooperation and ethical standards can be achieved only if all stakeholders seek to ensure human agency and oversight, and respect the established principles of privacy ~~and data governance~~, data governance and data protection - specifically those enshrined in Regulation (EU) 2016/679 of the European Parliament and of the Council (GDPR) - , transparency and accountability;

Amendment 56 #

Kris PEETERS

Draft opinion
Paragraph 4

4. Is of the opinion that effective cross- border cooperation and ethical standards can be achieved only if all stakeholders ~~seek~~commit to ensure human agency and oversight, ~~and respect the established principles of~~technical robustness and safety, privacy and data governance, transparency, non-discrimination, social and environmental wellbeing, and accountability;

Amendment 57 #

Andor DELI

Draft opinion
Paragraph 4

Amendment 58 #

Kris PEETERS, Juan Ignacio ZOIDO ÁLVAREZ, Roberta METSOLA, Axel VOSS

Draft opinion
Paragraph 4

4. Is of the opinion that effective ~~cross- border~~ cooperation and ethical standards can be achieved only if all stakeholders seek to ensure human agency ~~and~~, oversight, ~~and respect the established principles of privacy and data governance, transparency and accountability~~understanding, and control at any time, and respect the established Union legal and ethical principles;

Amendment 59 #

Draft opinion
Paragraph 4 a (new)

4a. Warns that possible bias in artificial intelligence applications could lead to automated discrimination, which has to be avoided by design and application rules;

Amendment 6 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Citation -1 e (new)

– having regard to the Council of Europe’s Framework Convention for the Protection of National Minorities, Protocol No 12 to the Convention for the Protection of Human Rights and Fundamental Freedoms, and the European Charter for Regional or Minority Languages,

Amendment 60 #

Draft opinion
Paragraph 5

5. Calls for a horizontal approach, including technology-neutral standards that apply to all sectors in which AI could be employed; calls on the Union to promote ~~strong and transparent~~a debate on how best the public and private sectors may cooperat~~ion~~e and share knowledge~~-sharing between the public and private sectors to create best practic~~ to create best practices; recalls that artificial intelligence technologies would not exist without training data sets populated with data harvested from citizens and from public sources, and calls for the Union to urgently explore mechanisms for making privately-held data sets publicly and freely available, without prejudice to applicable data protection rules;

Amendment 61 #

Kris PEETERS, Roberta METSOLA, Axel VOSS

Draft opinion
Paragraph 5

5. Calls for a horizontal and future- oriented approach, including technology- neutral standards that apply to all sectors in which AI could be employed, complemented by a vertical approach with sector-specific standards were appropriate; strongly believes that an ethical framework should apply to anyone intending to develop or operate artificial intelligence applications in the EU; favours a binding EU-wide approach to avoid fragmentation; calls on the Union to promote strong and transparent cooperation and knowledge-sharing between the public and private sectors to create best practices;

Amendment 62 #

Draft opinion
Paragraph 5

5. Calls for a horizontal approach, including technology-neutral standards that apply to all sectors in which AI could be employed; calls on the Union to promote strong and transparent cooperation and knowledge-sharing between ~~the~~ public and private sector~~s to create best practice~~ actors, including research and development, civil society and individual experts, and to encourage the development and sharing of best practices, notably to facilitate the de-biasing of datasets, and combatting discrimination in artificial intelligence and automated decision making systems;

Amendment 63 #

Draft opinion
Paragraph 5

5. Calls for a ~~horizontal approach~~risk-based differentiated approach to regulating artificial intelligence, robotics and related technologies, including technology-neutral ethical standards that apply to ~~all sectors in which AI could be employed~~high-risk technologies and to sectors in which the use of AI can threaten fundamental rights and liberties; calls on the Union to promote ~~strong and transparent~~ cooperation and knowledge- sharing between the public and private sectors to create best practices and continuously identify high-risk applications of AI;

Amendment 64 #

Axel VOSS

Draft opinion
Paragraph 5

5. Calls for a horizontal and future- oriented approach, including technology- neutral standards that apply to all sectors in which AI could be employed; calls on the Union to promote strong and transparent cooperation and knowledge-sharing between the public and private sectors to create best practices;

Amendment 65 #

Draft opinion
Paragraph 5 a (new)

5a. Calls for establishing binding rules for companies to publish transparency reports including the existence, functionality, process, main criteria, the logic behind the datasets used and possible outcome of algorithmic systems and efforts to identify, prevent and mitigate discrimination in artificial intelligence and automated decision making systems in a timely, impartial, easily-readable, and accessible manner;

Amendment 66 #

Katarina BARLEY, Sylvie GUILLAUME, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Miriam DALLI, Elena YONCHEVA

Draft opinion
Paragraph 5 a (new)

5a. Promotes a European Agency for Artificial Intelligence, which ensures a European coordination of AI standards and regulations; this centralized agency develops common criteria for a European certificate of ethical compliance, which also takes the data used for algorithmic processes into account;

Amendment 67 #

Draft opinion
Paragraph 5 b (new)

5b. Recommends that, considering the grave fundamental rights consequences at stake with many AI systems, all potentially high- and medium-risk systems, and especially those potentially operating in sensitive contexts, must undergo mandatory ex ante human rights impact and risk assessments, which include an evaluation of the collective, societal, institutional governance implications the system poses, and outlining adequate steps to mitigate; this risk-based approach should follow transparent rules to establish legal certainty;

Amendment 68 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Elena YONCHEVA

Draft opinion
Paragraph 5 b (new)

5b. Promotes Corporate Digital Responsibility on a voluntary basis; the EU should support corporations, who by choice use digital technologies and AI ethically within their companies; the EU should encourage corporations to become proactive by establishing a platform for companies to share their experiences with ethical digitalization, as well as coordinating the actions and strategies of participating companies;

Amendment 69 #

Draft opinion
Paragraph 5 c (new)

5c. Stresses that future regulation on artificial intelligence should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, determined by ex-ante human rights impact assessments, taking into account the specific use context of the artificial intelligence system, as well as the presumed opaqueness of decision-making; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no special legal obligations beyond those already in place; algorithmic systems that may cause material or immaterial harm to an individual, violate rights and freedoms, affect an individual’s access to resources, or negatively impact their participation in society, including by automated discrimination, shall not be deemed to belong to the lowest risk category; acknowledges that, for some AI systems, safeguards to mitigate bias will not be sufficient and that the deployment of certain systems for particular purposes will necessarily perpetuate and compound existing inequalities, such that fundamental rights for marginalized groups are compromised; considers that such systems should be banned;

Amendment 7 #

Draft opinion
Citation -1 f (new)

– having regard to Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin1a (Racial Equality Directive), __________________ 1a OJ L 180, 19.7.2000, p. 22.

Amendment 70 #

Draft opinion
Paragraph 5 d (new)

5d. Notes that it is essential for the risk assessment documentation, the software documentation, the algorithms and data sets used to be fully accessible to independent analysis; stresses in this respect the importance of lawful reverse- engineering;

Amendment 71 #

Draft opinion
Paragraph 5 e (new)

5e. Calls for a requirement for developers and deployers to provide for the relevant documentation on the use and design instructions, including source code, development tools and data used by the system, to be made easily accessible through a mandatory legal deposit, where a risk assessment indicates it necessary to protect fundamental rights and interests; recommends that for vital and advanced medical appliances, independent trusted entities retain the means necessary to provide services, for example to persons carrying these appliances, such as maintenance, repairs, and enhancements, including software updates, especially in the case where such services are no longer carried out by the original supplier, to preserve human dignity, autonomy, and self-determination of the individual;

Amendment 72 #

Roberta METSOLA, Juan Ignacio ZOIDO ÁLVAREZ, Javier ZARZALEJOS

Draft opinion
Paragraph 6

6. Stresses that the ~~protection of networks of interconnected AI and robotics must prevent security breaches, cyber-attacks and the misuse of personal data~~development of artificial intelligence, robotics and related technologies must follow a state of the art security-by-design approach to minimise the vulnerability to security breaches, cyber-attacks and the misuse of personal data; highlights the important role timely security fixes and software updates play in this regard; recommends where such updates are no longer provided by the original supplier, and where a risk assessment so indicates, suppliers should be required to provide for the relevant documentation on the use and design instructions, including the source code, development tools and data used by the system, to be made easily accessible through a mandatory legal deposit;

Amendment 73 #

Draft opinion
Paragraph 6

6. Stresses that the protection of networks of interconnected AI and robotics must prevent security breaches, cyber- attacks and the misuse of personal data and that this will require the relevant agencies, bodies and institutions both at the European level and the national level to work in cooperation with end users of these technologies;

Amendment 74 #

Kris PEETERS, Juan Ignacio ZOIDO ÁLVAREZ, Roberta METSOLA, Axel VOSS

Draft opinion
Paragraph 6

6. Stresses that the protection of networks of interconnected AI and robotics must prevent security breaches, cyber- attacks and the misuse of personal data; stresses that AI and associated technologies should always be used under human supervision;

Amendment 75 #

Draft opinion
Paragraph 6

6. Stresses that the protection of networks of interconnected AI and robotics must prevent security breaches, data leaks, data poisoning, cyber- attacks and the misuse of personal data; believes this will require a stronger cooperation between national and EU authorities;

Amendment 76 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Miriam DALLI, Elena YONCHEVA

Draft opinion
Paragraph 6

6. Stresses that the protection of networks of interconnected AI and robotics ~~must~~is important, and strong measures must be taken to prevent security breaches, cyber- attacks and the misuse of personal data;

Amendment 77 #

Draft opinion
Paragraph 6

6. Stresses that the ~~protection of networks of interconnected~~development of AI and robotics must prevent security breaches, cyber-attacks and the misuse of personal data;

Amendment 78 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Birgit SIPPEL, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Elena YONCHEVA

Draft opinion
Paragraph 6 a (new)

6a. Calls for a comprehensive risk assessment of AI, robotics and related technologies in addition to the impact assessment provided by Article 35 GDPR (Article 27 of Directive (EU) 2016/680 and Article 39 of Regulation (EU) 2018/1725); the more impact an algorithm has, the more transparency, auditability, accountability and regulation is needed; where an algorithmic decision leads to a limitation of fundamental rights, there needs to be a very robust assessment in place; in highly critical fields - when health, freedom or human autonomy are directly endangered - the implementation of AI should be prohibited;

Amendment 79 #

Draft opinion
Paragraph 6 a (new)

6a. Notes in this regard the provisions laid down in Regulation 2019/881 of the European Parliament and of the Council on ENISA and the Cyber Security Act, particularly ENISA's role in promoting public awareness and education campaigns directed at end users including on potential cyber threats and criminal activities online, and in promoting essential data protection measures; acknowledges the added value of this EU agency in this regard;

Amendment 8 #

Draft opinion
Citation -1 g (new)

– having regard to Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation1a (Equal Treatment in Employment Directive), __________________ 1a OJ L 303, 2.12.2000, p. 16.

Amendment 80 #

Draft opinion
Paragraph 6 a (new)

6a. Stresses that the malicious use of AI can pose a risk to the values of our democracies and the fundamental rights of the citizens of the European Union. Calls on the Commission to propose a framework that penalises those who, using this technology, distort the perception of reality through disinformation campaigns, or who provoke cyber-attacks in order to violate digital cyber-security.

Amendment 81 #

Draft opinion
Paragraph 6 a (new)

6a. Observes that data production and use resulting from the development, deployment and use of artificial intelligence, robotics and related technologies are rapidly increasing, thereby underlining the need to respect and enforce the rights of citizens to privacy and protection of personal data in line with Union law;

Amendment 82 #

Draft opinion
Paragraph 6 b (new)

6b. Points out that the possibility provided by these technologies of using personal and non-personal data to categorise and micro-target people, identify vulnerabilities of individuals, or exploit accurate predictive knowledge, has to be counterweighted by effectively enforced data protection and privacy principles such as data minimisation, the right to object to profiling and to control one’s data, the right to obtain an explanation of a decision based on automated processing, and privacy by design, as well as those of proportionality, necessity and limitation based on strictly identified purpose; points out that while certain models of predictive policing are more privacy-friendly than others, such as where probabilistic predictions are made about places or events and not about individual persons, predictive policing systems have proven to exacerbate overpolicing on the basis of existing bias such as racial profiling, or on migrant or working class backgrounds even where this does not correspond to actual crime levels;

Amendment 83 #

Draft opinion
Paragraph 6 b (new)

6b. Calls on the Commission and on the Council to provide all the agencies of the European Union with a regulatory framework in the use of AI technology that enables them to have a robust and effective cooperation with the public and private sectors for the protection of citizens, when breaches of security and misuse of personal data occur.

Amendment 84 #

Draft opinion
Paragraph 6 c (new)

6c. Reminds all law enforcement actors that use data processing and AI that Directive (EU) 2016/680 governs the processing of personal data by Member States for law enforcement purposes; requires that the collection and processing of personal data for law enforcement purposes must always be adequate, relevant and not excessive in relation to the specified, explicit and legitimate purposes for which they are processed; states that the purpose of and need for the collection of these data must be clearly proven; states that any decision based solely on automated processing, including profiling, which produces an adverse legal effect on the data subject or significantly affects him or her, is prohibited, unless authorised by Union or Member State law to which the controller is subject and which provides appropriate safeguards for the rights and freedoms of the data subject, at least the right to obtain human intervention on the part of the controllers; calls on the Commission, the European Data Protection Board and other independent supervisory authorities to issue guidelines, recommendations and best practices in order to further specify the criteria and conditions for decisions based on profiling and the use of AI for law enforcement purposes;

Amendment 85 #

Draft opinion
Paragraph 6 d (new)

6d. Stresses the importance of compliance with Directive (EU) 2016/680 as regards the carrying out of prior impact assessments and audits that take account of ethical concerns in order to assess the inclusiveness, accuracy and quality of data, and to ensure that individuals targeted by the decisions and/or actors involved in the decision- making processes are able to understand and challenge the collection or analysis, patterns and correlations and to prevent any harmful effects on certain groups of individuals;

Amendment 86 #

Draft opinion
Paragraph 6 e (new)

6e. Recommends that the Commission demonstrate it has clearly reviewed, assessed and adjusted its coordinated plan on AI in order to address the severe fundamental rights implications of Artificial Intelligence, and outline how such risks will be mitigated in the EU’s legislative approach and in the implementation of Member State national strategies;

Amendment 87 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology in the area of law enforcement and border control could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessHighlights that artificial intelligence, automated decision making and related technologies for purposes of mass surveillance, including using biometric technologies, by public and private actors in publicly accessible spaces are incompatible with existing Union law and fundamental rights; calls on the Commission, other EU institutions, and Member States to ban them in law and practice; further calls on the Commission to propose a ban on the use of AI to solely determine access to or delivery of essential public goods or services such as social security, policing, migration control; further, to propose a ban on uses of AI which purport to identify, analyse and assess emotion, mood, behaviour, and other sensitive identity traits, such as ethnicity or disability, in the delivery of essential services; further calls to propose a ban of autonomous lethal weapon systems and the use of AI for other uses which identify targets for lethal force, such as in law and immigration enforcement; and to propose a ban on general purpose scoring of residents, including mass-scale citizen scoring; notes that automated decision making systems in the area of law enforcement and border control must be carried out in accordance with Union law and with due regard to human rights and dignity, autonomy and self-determination of the individual, and the fundamental rights set out in the Charter; highlights that criteria and limits to such systems should be subjected to judicial review and submitted to democratic oversight and debate involving civil society; stresses that decisions by public authorities in the area of freedom, security and justice, such as getting bail or probation, being released from prison during a pandemic, being allowed to cross a border, getting asylum or international protection, or being heard in court, must never be taken by automated decision-making systems, but always involve meaningful assessment and judgement of a human, also to assign responsibility;

Amendment 88 #

Kris PEETERS

Draft opinion
Paragraph 7

7. Notes that AI ~~and~~, robotics and related technologyies in the area of law enforcement and border control could enhance public safety and security~~; stresses that its use must respect the principles of proportionality and necessity;~~ , yet also bear significant ethical risks that must be adequately addressed; stresses, therefore, that its use must respect the principles of proportionality and necessity and guarantee the respect of fundamental rights; supports the aim for maximum transparency, both with regards to the risk assessment of individual applications, as well as a general overview of the use of AI, robotics and related technologies in the area of law enforcement and border control; believes that EU agencies, especially in the field of Justice and Home Affairs, should be properly funded to research the effectiveness of such measures and should be equipped with the latest technologies;

Amendment 89 #

Draft opinion
Paragraph 7

7. ~~Notes that AI and robotic~~Believes that certain uses of AI cannot be considered as ethical as such, and that there are areas where any legal and ethical framework would not prevent risks of fundamentals rights violations; recalls that the use of AI, robotics and related technologyies in the area of law enforcement and border control ~~could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity~~poses extremely serious risks to fundamental rights; calls for a complete ban on the use of AI, robotics and related technologies in this arena; calls also for a ban on the use of facial recognition technology in public areas and a ban on affect recognition AI in any arena;

Amendment 9 #

Roberta METSOLA, Juan Ignacio ZOIDO ÁLVAREZ, Javier ZARZALEJOS, Kris PEETERS

Draft opinion
Citation -1 h (new)

– having regard to Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation)1a (GDPR), and to Directive (EU) 2016/680 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data by competent authorities for the purposes of the prevention, investigation, detection or prosecution of criminal offences or the execution of criminal penalties, and on the free movement of such data, and repealing Council Framework Decision 2008/977/JHA1b, __________________ 1a OJ L 119, 4.5.2016, p. 1. 1b OJ L 119, 4.5.2016, p. 89.

Amendment 90 #

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology in the area of law enforcement and border control could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity; considers that it should be possible for EU agencies in the field of Justice and Home Affairs to be equipped with the latest AI and robotic technologies, especially for the purposes of law enforcement and border control, and that this should be taken into account in the yearly budgets for the JHA agencies throughout the next MFF period (2021-2027);

Amendment 91 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Miriam DALLI, Elena YONCHEVA

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology are used more and more in the area of law enforcement and border control ~~could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity;~~ , often with adverse effects on individuals when it comes to their rights to privacy, data protection and non- discrimination; stresses that the deployment and use of these technologies must respect the principles of proportionality and necessity, the Charter of Fundamental Rights, in particular the rights to data protection, privacy and non- discrimination, as well as the relevant secondary Union law such as EU data protection rules;

Amendment 92 #

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology in the area of law enforcement and border control could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity; stresses that AI should never replace humans in issuing judgments, and should only be used in the justice system for the analysis and gathering of evidence;

Amendment 93 #

Clare DALY, Mick WALLACE, Konstantinos ARVANITIS, Cornelia ERNST, Pernando BARRENA ARZA, Anne-Sophie PELLETIER

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology in the area of law enforcement and border control could enhance public safety and security but also needs extensive and rigorous public scrutiny; stresses that its use must respect the principles of proportionality and necessity;

Amendment 94 #

Jana TOOM

Draft opinion
Paragraph 7

7. Notes that AI and robotic technology in the area of ~~law enforcement and~~ border control could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity and fully respect fundamental rights;

Amendment 95 #

Draft opinion
Paragraph 7 a (new)

7a. Notes the increasing use of AI- enabled labour-management systems; emphasises that the introduction of such systems raises significant questions about worker rights and safety; notes that AI systems used for worker control and management are inevitably optimised to produce benefits for employers, often at great cost to workers; recalls that Article 22 GDPR is not sufficient to adequately protect workers in the context of AI- enabled management systems; calls for urgent and specific regulation in this arena;

Amendment 96 #

Draft opinion
Paragraph 7 a (new)

7a. Calls on the Commission to thoroughly asses, in the context of its evaluation report pursuant to Article 92(5) of the ETIAS Regulation1a, if the feeding back of rejected travel authorisations into the risk score for future applicants from the same third country has created an unjustified bias against travellers that would have received a travel authorisation if they had applied from another third country, all other things being equal; __________________ 1aRegulation (EU) 2018/1240 of the European Parliament and of the Council of 12 September 2018 establishing a European Travel Information and Authorisation System (ETIAS), OJ L 236, 19.9.2018, p. 1–71, https://eur- lex.europa.eu/legal- content/EN/TXT/?uri=CELEX%3A32018 R1240

Amendment 97 #

Draft opinion
Paragraph 7 a (new)

7a. Calls on the Commission to enact strong and mandatory safeguards to prevent states and public authorities from misusing artificial intelligence, robotics and related technologies; stresses that the misuse of such technologies by authorities can become a direct threat to democracy and to the fundamental rights of our citizens;

Amendment 98 #

Katarina BARLEY, Marina KALJURAND, Sylvie GUILLAUME, Tudor CIUHODARU, Paul TANG, Petar VITANOV, Miriam DALLI, Elena YONCHEVA, Birgit SIPPEL

Draft opinion
Paragraph 8

8. Stresses that AI and robotics are not immune from making mistakes and can easily have inherent bias; notes that biases can be inherent in the underlying datasets, especially when historical data is being used, introduced by the developers of the algorithms, or generated when the systems are implemented in the real world setting; considers the need for legislators to reflect upon the complex issue of liability in the context of criminal justice.

Amendment 99 #

Draft opinion
Paragraph 8

8. Stresses that ~~AI and robotic~~individuals have a right to trust technology they use to perform in a reasonable manner, to respect their trust, and protect their good faith in it; points out that artificial intelligence, robotics and related technologies are not immune from making mistakes; considers the need for legislators to reflect upon the complex issue of liability in the context of criminal justice.

source: 653.783

2020/06/24 EMPL 127 amendments...

Amendment 1 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital -A (new)

-A. Having regard to the United Nations Guiding Principles on Business and Human Rights;

Amendment 10 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Recital A c (new)

Ac. whereas according to CEDEFOP’s European Skills and Jobs survey about 43% of EU adult employees have experienced new technologies at work, such as the introduction of machines and ICT systems; whereas about seven in ten EU workers require at least moderate digital skills to do their job;

Amendment 100 #

Draft opinion
Paragraph 5

5. Calls for the application of the precautionary principle with regard to new technologies based on AI; underlines that humans must always be in control of machines and AI and that AI decisions must be ~~reversible~~contestable and reversible; stresses that safety and security standards for AI must be respected and the importance of regular checks and controls in this regard to prevent erroneous AI output;

Amendment 101 #

Petra DE SUTTER, Lina GÁLVEZ

Draft opinion
Paragraph 5

5. Calls for the application of the precautionary principle with regard to new technologies based on AI; underlines that humans must always be in control of AI and that AI decisions must be reversible and employers must always prepare workers to work with co-bots or use new technologies based on AI safely and take the final responsibility if they fail to do so; emphasises in this regard that the rules on liability must be clearly defined both in the event of accidents at work and in the event of damage caused to third parties;

Amendment 102 #

Draft opinion
Paragraph 5

5. Calls for the application of the precautionary principle with regard to new technologies based on AI; underlines that humans must always be in control of AI and that AI decisions must be reversible; stresses that applicants must be duly informed in writing in case AI is used in the course of recruitment procedures and how in this case a human review can be requested in view of reversing an automated decision;

Amendment 103 #

France JAMET

Draft opinion
Paragraph 5

Amendment 104 #

Draft opinion
Paragraph 5

5. Calls for the application of prudence and the precautionary principle with regard to new technologies based on AI; underlines that humans must always manage, be in control of, and direct AI and that AI decisions must be reversible;

Amendment 105 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 5 a (new)

5a. Calls on the Commission to work closely with Member States on the design, implementation and enforcement of European ethical and safety standards for AI; notes that the EU has the potential to become a global leader in promoting a socially responsible approach to this technology and its use;

Amendment 106 #

Draft opinion
Paragraph 5 a (new)

5a. Stresses that applicants must be duly informed in writing in case AI is used in the course of recruitment procedures and how in this case a human review can be requested in view of reversing an automated decision;

Amendment 107 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 5 a (new)

5a. Stresses the need to carefully preserve the direct relations between employers, professionals, users and client in sensitive sectors;

Amendment 108 #

Marie-Pierre VEDRENNE, Sylvie BRUNET, Dragoş PÎSLARU, Stéphane BIJOUX, Véronique TRILLET-LENOIR, Ilana CICUREL

Draft opinion
Paragraph 6

6. Calls for a ~~legislative framework on~~n updated and appropriate legislative framework to address the new challenges concerning the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions; is of the view that special attention should be paid to all the new forms of work stemming from the implementation of new digital technologies, as in the case of the gig economy and digital platform workers, for whom it is vital to establish European reference legislation as soon as possible; recommends, in this regard, that priority be given to the definition of new rights, such as the right to disconnect, which are needed for all the new types of ‘ubiquitous’ work;

Amendment 109 #

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions; stresses also the need for this legislative framework to be based on the European acquis as regards data protection, in particular the General Data Protection Regulation (Regulation (EU) 2016/679);

Amendment 11 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Recital A d (new)

Ad. whereas on average, about one quarter of EU citizens have no or low- level digital skills; whereas the digital divide also has specific gender, accessibility, age and geographic aspects, which must be addressed;

Amendment 110 #

Cindy FRANSSEN, Loucas FOURLAS, Dennis RADTKE, Ádám KÓSA, Andrea BOCSKOR, Anne SANDER, Jarosław DUDA, Sara SKYTTEDAL, Krzysztof HETMAN

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions; points out the necessity for this legislative framework to be in line with the rules of protection of personal data and respect privacy of the workers;

Amendment 111 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE, Radka MAXOVÁ

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions, in line with a proportionate and risk-based approach such as outlined in the Commission White Paper on Artificial Intelligence;

Amendment 112 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 6

6. Calls for any European legislative framework on the ethical aspects of ~~AI applications in the workplace~~artificial intelligence and applications, robotics and related technologies to address the world of work, especially with regard to workers’ rights and working conditions;

Amendment 113 #

Johan DANIELSSON, Heléne FRITZON

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AIthe use, development and applications of AI in the workplace, especially with regard to workers’ rights, dignity and working conditions;

Amendment 114 #

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions; while respecting the autonomy of the social partners and collective agreements;

Amendment 115 #

Draft opinion
Paragraph 6

6. ~~Calls for a~~Stresses that data does not stop at national borders, and calls, therefore, for a European legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions;

Amendment 116 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 6

6. Calls for a legislative framework on the ethical aspects of AI applications in the workplace, especially with regard to workers’ rights and working conditions, with special regards to the rights of the platform workers;

Amendment 117 #

Draft opinion
Paragraph 6 – subparagraph 1a (new)

Stresses that the Skills Agenda for Europe must address the challenges of adapting and acquiring skills and knowledge, in view of the ecological and digital transition, including ethical aspects of AI, robotics and related technologies; underlines the need to make ethical aspects of AI an integral part of any education and training curricular for developers and people working with AI; considers it likewise important to ensure the comprehensive information of end users and consumers in this regard;

Amendment 118 #

Johan DANIELSSON, Heléne FRITZON

Draft opinion
Paragraph 6 a (new)

6a. Is of the opinion that in order better to manage the transition to an increasingly systematic use of AI in production processes, provision needs to be made, as soon as possible, for educational and training courses that take into account possible developments in employment and that are able to redirect workers’ skills towards sectors using little AI and robotisation (i.e. re-skilling and up-skilling), limiting the impact on employment levels and ensuring quality jobs and wages;

Amendment 119 #

Draft opinion
Paragraph 6 a (new)

6a. Highlights the need for competence development for workers and their representatives within the field of AI in the workplace, as crucial for improving decision-making and for an inclusive work environment;

Amendment 12 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital B

B. Wwhereas ~~AI offers economic and societal benefits, while at the same time raising a number of challenges;~~ new technologies in the workplace, like cloud computing, AI, Internet of Things and 5G hold great potential for better labour market functioning, offer economic and societal benefits, promote inclusive labour market, while at the same time carry new risks and raise a number of challenges, as recently showed during the Covid-19 pandemic and the consequent increase in teleworking, which posed a significant challenge to the entire IT infrastructure of companies or institutions;

Amendment 120 #

Draft opinion
Paragraph 6 a (new)

6a. Stresses that developers, programmers, decision-makers and companies providing AI and robotics solutions must be aware of their ethical responsibility when it comes to providing AI and robotics solutions;

Amendment 121 #

Draft opinion
Paragraph 6 a (new)

6a. Calls for new technologies to be integrated with humanist anthropological knowledge allowing the ethics of innovation to be discerned, so as to protect us from the dehumanisation of society;

Amendment 122 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 6 a (new)

6a. Highlights that with the help of AI-based technologies the skills and competences of people with disabilities can be better exploited in order to increase their level of employment;

Amendment 123 #

Draft opinion
Paragraph 6 b (new)

6b. Underlines that AI must not in any way affect the exercise of fundamental rights as recognised in the Member States and at Union level, including the right or freedom to strike or to take other action covered by the specific industrial relations systems in Member States, in accordance with national law and/or practice, nor affect the right to negotiate, to conclude and enforce collective agreements, or to take collective action in accordance with national law and/or practice;

Amendment 124 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 6 b (new)

6b. Stresses that AI- based technologies at the workplace, including applications used for telework, must be accessible for all, based on the design for all principles;

Amendment 125 #

Draft opinion
Paragraph 6 c (new)

6c. Calls for a legislative framework on the ethical aspects of AI applications at the workplace, in order to ensure a level- playing field and fair competition in the Single Market, especially with regard to workers’ rights as well as working conditions and the protection from disproportionate and illegal surveillance; calls on the Commission to consult with social partners and other stakeholders such as researchers and developers in this regard as well to explore the potential of digital technology and AI to increase the well-being of the workforce, including a better allocation of tasks, augmented competence development and work capacities as well as the reduction of exposure to harmful working conditions;

Amendment 126 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 6 d (new)

6d. Underlines that special attention must be payed to data collected at the workplace with the help of AI in so far as it is used for human resources decisions; calls on the Commission to analyse the need for special provisions on data protection at the workplace in the context of AI; stresses that workers must be the owners of their data;

Amendment 127 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 6 e (new)

6e. Underlines that AI must not reinforce gender inequalities and stereotypes by transforming analogue biases and prejudices into digital ones through algorithms;

Amendment 13 #

Lina GÁLVEZ, Alicia HOMS GINEL, Estrella DURÁ FERRANDIS, Brando BENIFEI, Pierfrancesco MAJORINO, Agnes JONGERIUS, Elisabetta GUALMINI, Milan BRGLEZ

Draft opinion
Recital B

B. Wwhereas AI should offers economic and societal benefits, as well as new job opportunities while at the same time raising a number of challenges;

Amendment 14 #

Draft opinion
Recital B

B. Wwhereas AI potentially offers economic and societal benefits, ~~while at the same time raising a number of challenges~~for example with regard to fighting inequalities and improving the quality of life, while at the same time raising a number of challenges; whereas the application of AI at the workplace can help to improve occupational health and safety, while it can also be used to monitor, evaluate, predict and steer the performance of workers with direct and implicit consequences for their careers and on their mental well-being;

Amendment 15 #

Cindy FRANSSEN, Loucas FOURLAS, Dennis RADTKE, Ádám KÓSA, Andrea BOCSKOR, Anne SANDER, Jarosław DUDA, Stelios KYMPOUROPOULOS, Krzysztof HETMAN

Draft opinion
Recital B

B. Wwhereas AI offers economic and societal benefits, while at the same time raising a number of challenges; Whereas AI offers economic and societal benefits, while at same time raising a number of challenges and that new AI applications should not have a negative impact on the current labour conditions;

Amendment 16 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Véronique TRILLET-LENOIR

Draft opinion
Recital B

B. Wwhereas AI offers economic and societal benefits and opportunities for both businesses and workers, while at the same time raising a number of challenges;

Amendment 17 #

Draft opinion
Recital B

B. Wwhereas AI offers economic and societal benefits, while at the same time raising a number of ethical, economic, social, employment-related and legal challenges;

Amendment 18 #

Draft opinion
Recital B a (new)

Ba. whereas a European approach to artificial intelligence should address, as a matter of priority, the ethical aspects and dilemmas associated with AI to ensure that it is human-centric, enhances human well-being, a sense of security, the well- being of society and the environment, and fully respects EU fundamental rights and values;

Amendment 19 #

Draft opinion
Recital B a (new)

Ba. whereas the use of artificial intelligence, automation, robotics and related technologies should be guided by the respect for human rights as a fundamental value, including in the field of work;

Amendment 2 #

Draft opinion
Recital A

A. Wwhereas the application of Artificial Intelligence (AI) in everyday life and in the workplace is increasing, the Covid-19 crisis has also proved that telework and digital working platforms are indispensable in the future of work and that all telework platforms must be accessible for people with disabilities as well;

Amendment 20 #

Miriam LEXMANN

Draft opinion
Recital B a (new)

Ba. whereas value of work efficiency must not lead to dehumanised digital future;

Amendment 21 #

Draft opinion
Recital B b (new)

Bb. whereas the full potential of AI can only be exploited if users are aware of the potential benefits and challenges that this technology brings; whereas it is necessary to incorporate this issue into the education process, including in terms of combating digital exclusion, and to conduct information campaigns at European level that give an accurate representation of all aspects of AI development;

Amendment 22 #

Draft opinion
Recital C

C. Whereas AI solutions and robotics C. are expected to have a strong impact on the labour market1 ~~and increase~~, gradually replacing humans in all repetitive activities, creating human-machine collaborative working systems and increasing the need for specializsed labour2.; __________________ 1 STOA, “‘The ethics of artificial intelligence: issues and initiatives”’ March 2020 2European Parliament, ‘Encouraging STEM Studies for the labour market’ March 2015

Amendment 23 #

Draft opinion
Recital C

C. Wwhereas AI solutions and robotics alre ~~expected to~~ady have a strong impact on the labour market1 and increase the need for specialized labour, therefore making upskilling and reskilling workers a must.2 __________________ 1 STOA, “The ethics of artificial intelligence: issues and initiatives” March 2020 2European Parliament “Encouraging STEM Studies for the labour market” March 2015

Amendment 24 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ

Draft opinion
Recital C

C. Wwhereas AI solutions and robotics are expected to have a ~~strong~~marked impact on the labour market1 and increase the need for skills and specialized labour.2 __________________ 1 STOA, “The ethics of artificial intelligence: issues and initiatives” March 2020 2European Parliament “Encouraging STEM Studies for the labour market” March 2015

Amendment 25 #

Draft opinion
Recital C

C. Wwhereas AI solutions and robotics are expected to have a strong impact on the labour market1 and increase the need for specialized labour2 and specific skills, especially digital skills; __________________ STOA, “The ethics of artificial intelligence: issues and initiatives” March 2020 European Parliament “Encouraging STEM Studies for the labour market” March 2015

Amendment 26 #

Draft opinion
Recital C

C. Wwhereas AI solutions and robotics are expected to have a strong impact on the labour market1 ~~and~~, being able to reduce the need for personnel in existing jobs, particularly where traditional and routine work which can be automated is concerned, and to increase the need for specialized labour2. __________________ 1STOA: “The ethics of artificial intelligence: issues and initiatives” March 2020. 2EuropeanParliament : “Encouraging STEM Studies for the labour market” March 2015.

Amendment 27 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C

C. Wwhereas AI solutions, automation, and robotics are expected to have a strong impact on the labour market1 and increase the need for specialized labour.2 __________________ 1 STOA, “The ethics of artificial intelligence: issues and initiatives” March 2020 2European Parliament “Encouraging STEM Studies for the labour market” March 2015

Amendment 28 #

Draft opinion
Recital C a (new)

Ca. whereas 42% of workers in companies that apply AI in their business processes believe that such activities lead to ethical issues, which must be addressed; whereas 28% of the employers believe that the application of AI has not developed at full scale because of a lack of ethical rules on this issue1a; _________________________ 1aCapgeminiResearch Institute, “Why addressing ethical questions in AI will benefit organizations”, July 2019

Amendment 29 #

Cindy FRANSSEN, Loucas FOURLAS, Stelios KYMPOUROPOULOS, Krzysztof HETMAN

Draft opinion
Recital C a (new)

Ca. whereas the introduction of AI has made the tasks of workers less executive and more complex, research1a has shown that AI leads to more ‘technostress’, with, in particular, less educated and elderly workers being more vulnerable to ‘technostress’. __________________ 1aMeylemans, Lise; Vanderstukken, Arne; Vereycken, Yennef; Ramioul, Monique; 2019. Aanwezigheid en impact van nieuwe technologieën in sectoren van ACV-CSC METEA.

Amendment 3 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Recital A

A. Wwhereas the application of Artificial Intelligence (AI), robotics and related technologies in everyday life and in the workplace is increasing;

Amendment 30 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE

Draft opinion
Recital C a (new)

Ca. whereas the employment landscape is rapidly evolving and it is estimated that 65% of today´s children entering primary school will ultimately end up working in completely new job types that still do not exist1a; __________________ 1aThe World Economic Forum, “The future of jobs” September 2018

Amendment 31 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C a (new)

Ca. whereas AI should aim to address a fair and just transition that targets the needs of workers in all kind of employment and economic sectors and ensures high-quality work, leaving no one behind;

Amendment 32 #

Draft opinion
Recital C b (new)

Cb. whereas the OECD has drawn up recommendations on artificial intelligence1a; __________________ 1aOECD, “Recommendation of the Council on Artificial Intelligence”, 2019, https://legalinstruments.oecd.org/en/instr uments/OECD-LEGAL-0449

Amendment 33 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE, Radka MAXOVÁ

Draft opinion
Recital C b (new)

Cb. whereas AI constitutes a strategic priority, considering that it should benefit citizens and society, by increasing competitiveness, creating job opportunities and economic prosperity;

Amendment 34 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C c (new)

Cc. whereas the Council of the European Union encourages the promotion of an ethical and human- centred approach with regard to AI1a; __________________ 1aCouncil of the European Union “Council Conclusions on Shaping Europe’s Digital future”, June 2020

Amendment 35 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE

Draft opinion
Recital C c (new)

Cc. whereas AI not only easily replaces jobs across indefinite fields, but also creates opportunities for positions that require more training and skills;

Amendment 36 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C d (new)

Cd. whereas there are no common provisions at Union level, as regards the application of AI at the workplace, which could lead to market distortions and competition disadvantages; whereas unified rules are necessary to provide workers and companies with a reliable regulatory framework;

Amendment 37 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C e (new)

Ce. whereas social partners at Union level concluded a framework agreement on digitalisation, which amongst others includes a chapter on “Artificial intelligence and guaranteeing the human in control principle” 1a; __________________ 1a European Social Partners Framework Agreement on Digitalisation, June 2020

Amendment 38 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C f (new)

Cf. whereas some Member States have already established special bodies to monitor and assess the influence of AI at the workplace;

Amendment 39 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Recital C g (new)

Cg. whereas workers and their representatives are often neither aware of AI applications nor the underlying functions and data; whereas lacking awareness and competitive pressure led to some companies applying AI in breach of existing regulations, such as data protection;

Amendment 4 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ, Estrella DURÁ FERRANDIS

Draft opinion
Recital A

A. Wwhereas the application of Artificial Intelligence (AI) in everyday life and ~~in the workplace is~~at work is constantly increasing;

Amendment 40 #

Draft opinion
Recital C h (new)

Ch. whereas efforts to tackle gender bias and inequality in the digital sector are insufficient; whereas the gender gap persists across all digital technology domains and especially with regard to AI, thereby solidifying a male-biased trajectory for the digital sector in the foreseeable future;

Amendment 41 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph -1 (new)

-1. Stresses the need for a single, shared, ethical and human-centred European approach to AI; points out that human beings, in particular, must be the drivers and beneficiaries of all innovation processes;

Amendment 42 #

Draft opinion
Paragraph 1

1. ~~Highlights the need to thoroughly assess the effects or implications of AI applications~~Stresses that the scope of any framework on the ethical aspects of artificial intelligence, robotics and related technologies should be adequate, proportionate and thoroughly assessed to avoid hampering future innovation and job creation; highlights in particular the need to thoroughly assess the opportunities and challenges that AI, robotics and related technologies can present in companies and in public administration, including in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that ~~workers and their representatives are consulted and receive sufficient information before AI applications are put to use;~~ the social partners are consulted on any European or national legislative or non-legislative frameworks regarding use of artificial intelligence, robotics and related technologies which impacts workplaces and invites them to include the ethical aspects of AI and its effects on the economy and employment, including in their joint work programmes at sectoral and cross-sectoral level where relevant;

Amendment 43 #

Stelios KYMPOUROPOULOS

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in ~~companies~~the private and in public ~~administration~~sector in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that ~~work~~consumers and ~~their~~workers representatives are consulted and receive sufficient information before AI applications are put to use;

Amendment 44 #

Cindy FRANSSEN, Loucas FOURLAS, Dennis RADTKE, Ádám KÓSA, Andrea BOCSKOR, Anne SANDER, Jarosław DUDA, Stelios KYMPOUROPOULOS, Sara SKYTTEDAL, Krzysztof HETMAN

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that workers and the~~ir representative~~ social partners are consulted and receive sufficient information before AI applications are put to use; considers it necessary that training be provided to all concerned employees when new AI technologies are introduced at the workplace;

Amendment 45 #

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that workers and their representatives ~~are consulted and~~ receive sufficient information before AI applications are put to use and that they actively participate in their design, development, and deployment, ensuring respect for human agency and personal rights;

Amendment 46 #

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess ~~the effects or~~in advance all the potential adverse effects and all the implications of AI applications in companies and in public administration in relation to workers, jobs, work organisation and workflows; considers it indispensable as part of this assessment that workers and their representatives are consulted in advance and receive sufficient information before AI applications are put to use;

Amendment 47 #

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that ~~workers and their representative~~the social partners are consulted and receive sufficient information before AI applications are put to use;

Amendment 48 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers including their impact on work-life balance, jobs and workflows; considers it indispensable as part of this assessment that workers and their representatives are consulted and receive sufficient information before AI applications are put to use; , taking into consideration that employees have the right to control their own data;

Amendment 49 #

France JAMET

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows and in the social security funding systems of the Member States; considers it indispensable as part of this assessment that workers and their representatives are consulted and receive sufficient information before AI applications are put to use;

Amendment 5 #

Draft opinion
Recital A a (new)

Aa. whereas Artificial Intelligence (AI) refers to systems that display intelligent behaviour by analysing their environment and taking actions – with some degree of autonomy – to achieve specific goals; whereas AI-based systems can be purely software-based, acting in the virtual world (e.g. voice assistants, image analysis software, search engines, speech and face recognition systems) or AI can be embedded in hardware devices (e.g. advanced robots, autonomous cars, drones or Internet of Things applications).1a __________________ 1aCommission Communication on Artificial Intelligence for Europe, COM(2018) 237 final

Amendment 50 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that workers and their representatives are consulted and receive sufficient information before AI applications are put to use; underlines that the deployment of AI needs to be transparent and that AI systems at the workplace must not undermine the privacy and dignity of workers;

Amendment 51 #

Johan DANIELSSON, Heléne FRITZON

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and workflows; considers it indispensable as part of this assessment that workers and their representatives are consulted and receive sufficient information from the start of the decision process and before AI applications are put to use;

Amendment 52 #

Draft opinion
Paragraph 1

1. Highlights the need to thoroughly assess the effects or implications of AI applications in companies and in public administration in relation to workers, jobs and work flows; considers it indispensable as part of this assessment that workers and their representatives are consulted and, as part of the social dialogue, receive sufficient information before AI applications are put to use;

Amendment 53 #

Miriam LEXMANN

Draft opinion
Paragraph 1 a (new)

1a. Considers human dignity a fundamental value and human control a fundamental principle of the revolutionary technological progress;

Amendment 54 #

Draft opinion
Paragraph 1 a (new)

1a. Acknowledges that data collection, management and analysis involve ethical considerations and challenges and may affect workers’ rights, such as the right to privacy; considers that the collection and management of workers’ data should be also protected by collective agreements and regulations, ensuring transparency, traceability and accountability, as well as control by employees of their own data - including in the case of termination of contracts- and the obligation of companies to be audited; also data analysis should be monitored regularly and systematically to prevent bias and direct and indirect discrimination and corrective measures should be ensured;

Amendment 55 #

Draft opinion
Paragraph 1 b (new)

1b. Calls for the establishment of an artificial intelligence governance body at company level, jointly comprised of representatives of workers and employers, in order to monitor the development, deployment and use of artificial intelligence, robotics and related technologies within companies; alternatively the duties and responsibilities of such a body could be added to existing dialogue structures, such as works councils;

Amendment 56 #

Draft opinion
Paragraph 1 c (new)

1c. Points out that a comprehensive risk assessment should come before the development, deployment and implementation of artificial intelligence, robotics and related technologies systems, evaluating its impact on fundamental rights and working conditions, including in terms of occupational health and safety, as well as its social consequences; assessments should include risks related to human decision-making and social discrimination, as well as the evaluation of occupational risks arising;

Amendment 57 #

Draft opinion
Paragraph 2

2. Points out that AI solutions have the ~~potential~~capacity to be very influential, something which can be used to improve working conditions and the quality of life, yet ~~they~~ can also ~~lead to disproportionate and illegal surveillance of workers, discriminatory treatment due to biased algorithms, including gender biased algorithms - and they~~have a negative impact on the company and its workers, and can undermine the freedom, the dignity and autonomy of people; ~~__________________ 3European employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020.~~Parliament: “Education and

Amendment 58 #

France JAMET

Draft opinion
Paragraph 2

Amendment 59 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life~~, yet they can also lead to disproportionate and illegal surveillance of workers,~~ for all workers particularly as regards employment and social inclusion of, and accessibility for, workers with disabilities as well as improving work-life balance, while they can also in the context of the workplace raise concerns as regards privacy and occupational health and safety such as the right to disconnect and discriminatory treatment or recruitment processes due to biased algorithms, including gender and racially biased algorithms3 ~~- and they can undermine~~ ; stresses the need to ensure that people from diverse backgrounds, including women, young people and people withe di~~gnity and autonomy of people~~sabilities are included in inclusive development, deployment and use of AI; __________________ 3European Parliament “Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020

Amendment 6 #

Draft opinion
Recital A a (new)

Aa. whereas artificial intelligence (AI) is a technology of the future and is expected to benefit citizens and society by improving the quality of life and creating new employment opportunities and more sustainable business models;

Amendment 60 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, yet they can also lead to disproportionate and illegal surveillance of workers, if misused, leading to discriminatory treatment due to biased algorithms~~, including gender biased algorithms3 -~~ and they can undermine the dignity and autonomy of people; ~~__________________ 3 European Parliament “Education and employment of women in sci~~at the same time points out that AI has the potential to prevent human bias in recruitment procedures, can support remote applications, including for those that face disabilities, matching people to jobs due to fact-based matching, identify and predict the interoperability of skills, traits and competences, ~~technology and the digital economy, including AI and its influence on gender equality”, April 2020~~support human resource management, and predict labour market developments, and to those faced with labour market change and challenge, responsible AI in human resources services will contribute to their labour market activation, transition, lifelong learning, and employability;

Amendment 61 #

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, yet they can also lead to disproportionate and illegal surveillance of workers, discriminatory treatment due to biased algorithms, including gender or ethnically biased algorithms3 ~~- and they can undermine the dignity and autonomy of peopl~~and algorithms that are to the detriment of vulnerable groups and that undermine the privacy and dignity of workers; is concerned, furthermore, that AI can undermine the dignity and autonomy of workers and contribute to mental health problems, such as burnout, occupational stress, psychological overload and fatigue; __________________ 3 European Parliament “Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020

Amendment 62 #

Petra DE SUTTER, Lina GÁLVEZ

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, yet they can also lead to disproportionate and illegal surveillance of workers, discriminatory treatment due to biased algorithms, including gender biased algorithms~~3 -~~ and they can undermine the dignity and autonomy of people; __________________ 3 European Parliament “Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020algorithms at the detriment of vulnerable groups, which can undermine the privacy and dignity of workers; is concerned, furthermore, that AI can undermine the dignity and autonomy of workers and contribute to mental health problems, such as burnout, occupational stress, psychological overload and fatigue;

Amendment 63 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE, Radka MAXOVÁ

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, ~~yet they can also~~and points out that these technologies remain subject to legislation on the use of data and privacy such as the General Data Protection Regulation, to avoid situations leading to disproportionate and illegal surveillance of workers, discriminatory treatment due to biased algorithms, including gender biased algorithms3 - and they can undermine the dignity and autonomy of people; __________________ 3 European Parliament “Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020

Amendment 64 #

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, yet they can also lead to disproportionate and illegal surveillance ~~of workers,~~and monitoring of workers, which can cause psychological stress and job insecurity, as well as to discriminatory treatment due to biased algorithms, including gender biased algorithms3 -, and they can undermine the dignity and autonomy of people, violate human rights and values and put in danger human diversity; __________________ 3 European Parliament “Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”, April 2020

Amendment 65 #

Cindy FRANSSEN, Loucas FOURLAS, Dennis RADTKE, Ádám KÓSA, Andrea BOCSKOR, Anne SANDER, Jarosław DUDA, Stelios KYMPOUROPOULOS, Sara SKYTTEDAL, Krzysztof HETMAN

Draft opinion
Paragraph 2

Amendment 66 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 2

2. Points out that AI solutions have the potential to improve working conditions and the quality of life, yet they can also lead to distortions in the organisation, supervision and monitoring of work; highlights the risk, in particular, of disproportionate and illegal surveillance of workers, and discriminatory treatment due to biased algorithms, including gender biased algorithms3 - ~~and they~~AI solutions, in addition, can undermine the dignity and autonomy of people; __________________ 3 European Parliament “‘Education and employment of women in science, technology and the digital economy, including AI and its influence on gender equality”’, April 2020

Amendment 67 #

Draft opinion
Paragraph 2 – subparagraph 1 (new)

Underlines that competent authorities should have access to all information concerning the data used for training, statistical models and theoretical principles related to AI solutions as well as the empirical validity of their outcomes;

Amendment 68 #

Draft opinion
Paragraph 2 a (new)

2a. Highlights the need to develop a robust certification system, based on test procedures underpinned by the precautionary principle, enabling businesses to attest that their AI products are reliable and safe, leading to greater benefits for the whole community;

Amendment 69 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 2 a (new)

2a. Stresses that, to protect workers from bias and direct and indirect discrimination as a result of AI decisions, whistleblowing mechanisms and monitoring and appeal procedures should be put in place, particularly in cases such as recruitment, promotion or dismissal, protecting workers from retaliation;

Amendment 7 #

Miriam LEXMANN

Draft opinion
Recital A a (new)

Aa. whereas AI will significantly transform current socio-economic structures and should be subject to an appropriate regulatory framework;

Amendment 70 #

Draft opinion
Paragraph 2 a (new)

2a. Recalls that AI and robotics should respect human dignity and autonomy, and calls for clear rules to avoid potential misuse such as disproportionate and illegal surveillance of workers, deviant use of the so-called “socio-genomics” or bypassing systems for social dialogue;

Amendment 71 #

Sara SKYTTEDAL

Draft opinion
Paragraph 2 a (new)

2a. Stresses the importance of creating an AI rulebook that safeguards both social aspects and the EU’s competitiveness;

Amendment 72 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to ~~receive all relevant information on the ethical aspects of AI applications so that they may make informed decisions~~be informed when a system uses AI and related technologies, particularly AI systems which personalise a product or service to its users, whether they can switch off or limit such personalisation and in cases where they are faced with an automated-decision making technology so that they may make informed decisions; stresses the need to ensure that professionals using algorithms understand how data is used and valued; considers that workers should receive adequate training and education on the various issues involved in the use and implication of the AI tools they use;

Amendment 73 #

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on the ethical aspects of AI applications so that they may make informed ~~decision~~and conscious decisions; stresses the need – in order to seize all the opportunities offered by the digital revolution – to launch large-scale IT literacy processes, with a particular focus on all the disadvantaged social groups at risk of social marginalisation owing to a lack of digital tools and skills;

Amendment 74 #

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on the ethical aspects of AI applications and their design process, so that they may make informed decisions; recommends to promote a space of dialogue between all involved parties, incorporating civil society, to raise awareness of the ethical challenges posed by artificial intelligence, robotics and related technologies;

Amendment 75 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on the ethical aspects of AI applications so that they may make informed decisions; draws attention to breaches of privacy generated by AI for the purpose of discrimination;

Amendment 76 #

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive meaningful and all relevant information on the use of AI applications that affect them, on the ethical aspects of AI applications so that they may make informed decisions;

Amendment 77 #

Stelios KYMPOUROPOULOS

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information, in accessible form, on the ethical aspects of AI applications so that they may make informed decisions;

Amendment 78 #

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on the ethical aspects of AI applications in easily understandable, accessible form so that they may make informed decisions;

Amendment 79 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on all the ethical aspects of AI applications so that they may make informed decisions;

Amendment 8 #

Sara SKYTTEDAL

Draft opinion
Recital A a (new)

Aa. whereas AI is an important tool for strengthening the EU’s competitiveness;

Amendment 80 #

Draft opinion
Paragraph 3

3. Considers that it should be mandatory for users and consumers to receive all relevant information on the ethical aspects of AI applications ~~so that they may m~~to take informed decisions;

Amendment 81 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 3 – subparagraph 1 a (new)

Underlines that trustworthy AI has to be fair, transparent, safe and secure and to comply with all applicable laws and regulations including the General Data Protection Regulation (GDPR) throughout the system’s entire life cycle, especially when it is deployed at the workplace;

Amendment 82 #

Draft opinion
Paragraph 3 a (new)

3a. Emphasises the need to address the question of what standards of labour competitiveness may be needed in the field of employment when existing human skills are artificially - or even genetically - developed or directly supplemented, creating an impenetrable advantage for those who have access to such tools and interventions, given that human dignity is and must be at the heart of the European and human rights approach;

Amendment 83 #

Dragoş PÎSLARU, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 3 a (new)

3a. Highlights the importance of ensuring that those involved in the design, development and marketing of AI act in accordance with ethical standards and social responsibility criteria; considers that training programmes should cover the understanding of ethics and the development of skills for ethical purposes, including the recognition of biases;

Amendment 84 #

Draft opinion
Paragraph 3 a (new)

3a. Notes that the opportunities of AI solutions rely on Big Data, with a need for a critical mass of data to train algorithms and refine results; welcomes in this regard the Commission proposed creation of a common Data Space in the EU to strengthen data exchange and support research in full respect of data protection for workers and employers;

Amendment 85 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE, Radka MAXOVÁ

Draft opinion
Paragraph 3 a (new)

3a. Considers that a European Teleworking Agenda must take into account the impact of AI and related technologies on remote work;

Amendment 86 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 3 a (new)

3a. Stresses that the development and spread of AI must not restrict access to services and benefits;

Amendment 87 #

Draft opinion
Paragraph 3 b (new)

3b. Highlights the importance of education and continuous learning to develop the skills of tomorrow, to ensure the complementarity between human and machine at work, and to ensure the re- skilling of workers in the sectors heavily impacted by the risk of automation; calls therefore for a Programme for European AI teaching posts, a Europe-wide academic AI exchange programme for university networks;

Amendment 88 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics ~~do not only benefit company~~ benefit society at large, including companies and workers by driving EU growth and competitiveness and the creation of quality employment; notes that sectors and occupations requiring medium-or lowner~~s and shareholders, but also society at~~-level skills are more prone to automation; considers it essential that the Member States, with the support of the Commission and in co- operation with relevant actors such as businesses and academia, invest in high quality, responsive and inclusive education, vocational training and lifelong lear~~ge, especially where such gains co~~ning systems and up- and re- skilling policies in order to equip students and workers with basic literacy, numeracy and digital skills as well as competences in sciences, technology, engineering and mathematics (STEM) and transversal soft skills such as entrepreneurship, creativity and critical thinking to avoid skills obsolescence and ensure employability in transformed a~~t the expense of jobs;~~ nd future-oriented sectors; calls for the Updated Skills Agenda for Europe and the reinforced MFF for 2021- 2027 to play a strong role in this regard;

Amendment 89 #

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics ~~do not only benefit company owners and shareholders, but also society at large, especially where such gains come at the expense of jobs~~serve the development and growth of undertakings, without causing a reduction in the number of jobs, and benefit society at large;

Amendment 9 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Abir AL-SAHLANI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Recital A b (new)

Ab. whereas AI, robotics and related technologies can make a huge contribution to achieving the common goal of improving the lives of citizens and fostering prosperity within the EU as well as managing the twin transitions towards a green and digital economy, if harnessed correctly;

Amendment 90 #

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics do not only benefit company owners and shareholders, but also society at large, especially where such gains come at the expense of jobs; points out, particularly, that employees should be able to participate in the redistribution of company profits through better salaries and career possibilities, investment in skills, reduction of working time or better work-life balance, and that collective bargaining and collective bargaining agreements are key to integrate social progress and high-quality jobs with an ethical approach to AI;

Amendment 91 #

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics do not only benefit company owners and shareholders, but also society at large, especially where such gains come at the expense of jobs and in this context stresses the need of upskilling and reskilling of workers which can help them use digital tools, keep their jobs or learn how to work with co-bots or with AI based technologies;

Amendment 92 #

Lina GÁLVEZ, Brando BENIFEI, Elisabetta GUALMINI, Pierfrancesco MAJORINO, Estrella DURÁ FERRANDIS, Alicia HOMS GINEL, Agnes JONGERIUS, Milan BRGLEZ

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics do not only benefit company owners and shareholders, but also society at large, especially where such gains come at the expense of jobs; stresses the crucial importance of digital skills and lifelong learning to help workers adapt to the changing needs of the labour market;

Amendment 93 #

Draft opinion
Paragraph 4

4. Stresses the need to ensure that productivity gains due to the development and use of AI and robotics do not only benefit company owners and shareholders, but also society at large, especially where such gains come at the expense of jobs; underlines that companies deploying AI have the responsibility to provide the necessary re- and upskilling for the employees concerned;

Amendment 94 #

Draft opinion
Paragraph 4

Amendment 95 #

Dragoş PÎSLARU, Sylvie BRUNET, Marie-Pierre VEDRENNE, Jordi CAÑAS, Ilana CICUREL, Samira RAFAELA, Atidzhe ALIEVA-VELI, Radka MAXOVÁ, Véronique TRILLET-LENOIR

Draft opinion
Paragraph 4 a (new)

4a. Calls on the Member States to address the impact of AI technologies, automation and robotics in terms of possible job losses and increased job insecurity and to adopt measures to ensure long-term sustainability of national social protection systems by reforming taxes and social contributions to compensate for smaller public revenues;

Amendment 96 #

Draft opinion
Paragraph 4 a (new)

4a. Reminds that the EU employment and social acquis remains fully applicable irrespective of the involvement of AI and believes the Commission and Member States should ensure it is enforced adequately where AI is used and address any potential legislative gaps;

Amendment 97 #

Jordi CAÑAS, Sylvie BRUNET, Samira RAFAELA, Marie-Pierre VEDRENNE, Radka MAXOVÁ

Draft opinion
Paragraph 4 a (new)

4a. Calls for the adoption of an updated Skills Agenda for Europe and Digital Education Action Plan in order to provide digital literacy to workers across all sectors;

Amendment 98 #