Activities of Sergey LAGODINSKY related to 2020/2012(INL)
Shadow reports (1)
REPORT with recommendations to the Commission on a framework of ethical aspects of artificial intelligence, robotics and related technologies
Shadow opinions (1)
OPINION with recommendations to the Commission on the framework of ethical aspects of artificial intelligence, robotics and related technologies
Legal basis opinions (0)
Amendments (219)
Amendment 1 #
Draft opinion
Citation -1 (new)
Citation -1 (new)
– having regard to Articles 2 and 3 of the Treaty on European Union (TEU),
Amendment 2 #
Motion for a resolution
Citation 2 a (new)
Citation 2 a (new)
- having regard to Article 169 of the Treaty on the Functioning of the European Union,
Amendment 2 #
Draft opinion
Citation -1 a (new)
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 3 #
Motion for a resolution
Citation 2 b (new)
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 3 #
Draft opinion
Citation -1 b (new)
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 4 #
Motion for a resolution
Citation 7 a (new)
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 4 #
Draft opinion
Citation -1 c (new)
Citation -1 c (new)
– having regard to Articles 21 and 22 of the EUCFR,
Amendment 5 #
Draft opinion
Citation -1 d (new)
Citation -1 d (new)
– having regard to the preamble to the TEU,
Amendment 6 #
Motion for a resolution
Citation 14 a (new)
Citation 14 a (new)
- having regard to the OECD Council Recommendation on Artificial Intelligence adopted on 22 May 2019,
Amendment 6 #
Draft opinion
Citation -1 e (new)
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 7 #
Motion for a resolution
Recital A
Recital A
A. whereas so-called ‘artificial intelligence’, or extended intelligence, robotics and related technologies withare 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 7 #
Draft opinion
Citation -1 f (new)
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 8 #
Draft opinion
Citation -1 g (new)
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 9 #
Draft opinion
Citation -1 h (new)
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 10 #
Draft opinion
Citation -1 i (new)
Citation -1 i (new)
Amendment 11 #
Draft opinion
Citation -1 j (new)
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 12 #
Draft opinion
Citation -1 k (new)
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)
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)
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)
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 #
Motion for a resolution
Recital B
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, and are advanced for the benefit of humanity;
Amendment 16 #
Draft opinion
Recital D (new)
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)
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)
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)
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 20 #
Motion for a resolution
Recital C
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 the worldrustworthy technology;
Amendment 20 #
Draft opinion
Paragraph -1 a (new)
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)
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 23 #
Motion for a resolution
Recital D
Recital D
Amendment 25 #
Draft opinion
Paragraph 1
Paragraph 1
1. Believes that any ethical and legal framework should seek to respect human autonomyrespect and preserve human dignity, autonomy, and self- determination of the individual, prevent harm, promote fairness, and respecimplement the principles of explicability, interpretability, auditability, as well as traceability, transparency, and accessibility of technologies;
Amendment 27 #
Motion for a resolution
Recital E
Recital E
E. whereas such questionlegal 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 that would bringo increase legal certainty tofor businesses and citizens alike and to secure civil rights, labour rights and environmental standards;
Amendment 31 #
Draft opinion
Paragraph 1 a (new)
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 37 #
Motion for a resolution
Recital F
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 38 #
Motion for a resolution
Recital G
Recital G
G. whereas thate Union’s legal framework should encompass all situations requiring due consideration of the Union’s principles and valurelevant stages, namely from the development, and deployment andto the use of the relevant technologies and their components;
Amendment 38 #
Draft opinion
Paragraph 2
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 41 #
Motion for a resolution
Recital H
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 conceptsof concepts that make up the technologies, such as algorithms, and software, data or biometric recognition;
Amendment 43 #
Draft opinion
Paragraph 3
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 45 #
Motion for a resolution
Recital I
Recital I
I. whereas action at Union level is justified by the need forto ensure a homogenous application of common ethical principles enshrined in law when developing, deploying and using artificial intelligence, robotics and related technologies;
Amendment 48 #
Motion for a resolution
Recital J
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 49 #
Motion for a resolution
Recital J a (new)
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 50 #
Motion for a resolution
Recital J b (new)
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 54 #
Motion for a resolution
Recital K
Recital K
K. whereas each Member State should destablish a nationalignate 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 58 #
Draft opinion
Paragraph 4
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 accountabilityunderstanding, and control at any time, and respect the established Union legal and ethical principles;
Amendment 62 #
Draft opinion
Paragraph 5
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 sectors 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 65 #
Draft opinion
Paragraph 5 a (new)
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 67 #
Draft opinion
Paragraph 5 b (new)
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 69 #
Draft opinion
Paragraph 5 c (new)
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 70 #
Draft opinion
Paragraph 5 d (new)
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)
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 #
Draft opinion
Paragraph 6
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 datadevelopment 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 74 #
Motion for a resolution
Paragraph 1
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 humunderstanding and control at any time;
Amendment 81 #
Draft opinion
Paragraph 6 a (new)
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)
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 84 #
Motion for a resolution
Paragraph 2
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 principlesto bear a risk should always follow from an impartial, regulated and external assessment;
Amendment 84 #
Draft opinion
Paragraph 6 c (new)
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)
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)
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 #
Draft opinion
Paragraph 7
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 89 #
Motion for a resolution
Subheading 4
Subheading 4
Safety features, transparency and, accountability, and accessibility
Amendment 90 #
Motion for a resolution
Paragraph -3 a (new)
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)
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)
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)
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
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 96 #
Draft opinion
Paragraph 7 a (new)
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 99 #
Draft opinion
Paragraph 8
Paragraph 8
8. Stresses that AI and roboticindividuals 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.
Amendment 101 #
Motion for a resolution
Paragraph 4
Paragraph 4
4. Underlines that explainability, is essential to ensuring thatnterpretability, auditability, traceability, and transparency, as well as access to technology, data and computing systems underlying such technologies contribute to citizens’ trust thesein technologiesy, 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 105 #
Draft opinion
Paragraph 8 a (new)
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 #
Motion for a resolution
Paragraph 4 a (new)
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)
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)
Subheading 4 a (new)
Redress
Amendment 109 #
Motion for a resolution
Paragraph 4 c (new)
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 111 #
Motion for a resolution
Paragraph 5
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 dignitythem, 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 #
Draft opinion
Paragraph 8 b (new)
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 116 #
Motion for a resolution
Paragraph 6
Paragraph 6
6. Affirms that possible bias in and discrimination by software, algorithms and data should be addressed by settingcan 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 wouldto have the potential to turn software, algorithms and data into a considerable counterbalance ton asset in combating bias and discrimination, and a positive force for positive social change;
Amendment 121 #
Motion for a resolution
Paragraph 6 a (new)
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 124 #
Motion for a resolution
Paragraph 7
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 137 #
Motion for a resolution
Paragraph 8
Paragraph 8
8. Proposes that theStresses 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 maximfor citizedns and explored through responsible research and innovation that requires thepublic good; calls for mobilisation of resources by the Union and its Member States in order to develop and support responsible innovation;
Amendment 140 #
Motion for a resolution
Paragraph 9
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 142 #
Motion for a resolution
Paragraph 10
Paragraph 10
10. States that it is essential that artificial intelligence, robotics and related technologies supportbenefit 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 minimisingcontribute 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 147 #
Motion for a resolution
Paragraph 11
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 innovationdevelopment that requires the mobilisation of resources by the Union and its Member States;
Amendment 148 #
Motion for a resolution
Paragraph 12
Paragraph 12
12. Highlights that the development, deployment and use of these technologies provide numerous opportunities for the achievementpromoting of the Sustainable Development Goals outlined by the United Nations, global energy transition and decarbonisation;
Amendment 151 #
Motion for a resolution
Paragraph 12 a (new)
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
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
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 theeffectively 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 156 #
Motion for a resolution
Paragraph 15
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 159 #
Motion for a resolution
Paragraph 15 a (new)
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 168 #
Motion for a resolution
Paragraph 16
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 171 #
Motion for a resolution
Paragraph 17
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 controllaw;
Amendment 179 #
Motion for a resolution
Paragraph 18
Paragraph 18
18. Underlines the need to ensure that data belonging toon, or stemming from vulnerable groups, such as people with disabilities, patients, children, minorities and migrants, are protected adequately protected;
Amendment 180 #
Motion for a resolution
Paragraph 18 a (new)
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 183 #
Motion for a resolution
Subheading 10
Subheading 10
Amendment 185 #
Motion for a resolution
Paragraph 19
Paragraph 19
19. Notes the added value of having nationalat 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 197 #
Motion for a resolution
Paragraph 21
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 218 #
Motion for a resolution
Paragraph 23
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 practionsce of each nationalthe competent supervisory authority as referred to in the previous sub-sectionies, and of contributing to the collaboration of research and development in the field throughout the EU;
Amendment 228 #
Motion for a resolution
Paragraph 24
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 technologiesy;
Amendment 231 #
Motion for a resolution
Paragraph 24 a (new)
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 241 #
Motion for a resolution
Paragraph 25
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 nationalcompetent supervisory authority;
Amendment 242 #
Motion for a resolution
Paragraph 25 a (new)
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)
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)
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 246 #
Motion for a resolution
Paragraph 26
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 differentcontinuing the critical and ethics-based dialogue with third countries implementing alternative models of AI regulation, development and deployment models.
Amendment 250 #
Motion for a resolution
Paragraph 27
Paragraph 27
27. Recalls that the opportunities and risks inherent to these technologies have a global dimension that requires a consistent approach at intinternational coopernational level and thus calls on the Commission to work in bilateral and multilateral settings to advocate and ensure ethical compliance.
Amendment 260 #
Motion for a resolution
Paragraph 29
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 resultingenshrined in a legal framework at Union level, supervisported 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 266 #
Motion for a resolution
Paragraph 30
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 this report and the annex hereto;
Amendment 278 #
Motion for a resolution
Annex I – part A – point I – indent 1
Annex I – part A – point I – indent 1
- to build trust in artificial intelligence, robotics and relatedintelligent 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 281 #
Motion for a resolution
Annex I – part A – point I – indent 2
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 procesinnovation and business development process, and during the subsequent phase of use by professionals and private individuals;
Amendment 285 #
Motion for a resolution
Annex I – part A – point I – indent 3
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 289 #
Motion for a resolution
Annex I – part A – point I – indent 4
Annex I – part A – point I – indent 4
Amendment 291 #
Motion for a resolution
Annex I – part A – point I – indent 5
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 298 #
Motion for a resolution
Annex I – part A – point II – indent 4
Annex I – part A – point II – indent 4
- the work carried out by the “Scompetent supervisory Aauthority”ies in eachthe Member States to ensure that ethical principles are applied to artificial intelligence, robotics and related technologies;
Amendment 300 #
Motion for a resolution
Annex I – part A – point II – indent 5
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 307 #
Motion for a resolution
Annex I – part A – point III – indent 3
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
Annex I – part A – point III – indent 4
- safeguards and remedies against bias and discrimination;
Amendment 311 #
Motion for a resolution
Annex I – part A – point III – indent 5
Annex I – part A – point III – indent 5
- social responsibility and gender balanceequality in artificial intelligence, robotics and related technologies;
Amendment 313 #
Motion for a resolution
Annex I – part A – point III – indent 7
Annex I – part A – point III – indent 7
- respect for fundamental rights, including privacy and limitations to the use of biometric recognition;
Amendment 316 #
Motion for a resolution
Annex I – part A – point IV – introductory part
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 320 #
Motion for a resolution
Annex I – part A – point IV – indent 2
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 329 #
Motion for a resolution
Annex I – part A – point V – indent 1
Annex I – part A – point V – indent 1
- to supervise the application of the proposed Regulationrelevant Union law;
Amendment 331 #
Motion for a resolution
Annex I – part A – point V – indent 2
Annex I – part A – point V – indent 2
- to issue guidance as regards the application of the proposed Regulationrelevant Union law;
Amendment 333 #
Motion for a resolution
Annex I – part A – point V – indent 3
Annex I – part A – point V – indent 3
- to liaise with the “SMember States and concerned competent supervisory Aauthority” in each Member State and coordinate their mandate and tasksies, coordinate their contributions and collaboration;
Amendment 336 #
Motion for a resolution
Annex I – part A – point V – indent 4
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 338 #
Motion for a resolution
Annex I – part A – point V – indent 5
Annex I – part A – point V – indent 5
- to support regular exchanges with concerned stakeholders and the civil societycross-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)
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 340 #
Motion for a resolution
Annex I – part A – point V – indent 5 b (new)
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
Annex I – part A – point VI – introductory part
VI. The key tasks of the “Scompetent supervisory Aauthority”ies in eachthe Member States should be:
Amendment 344 #
Motion for a resolution
Annex I – part A – point VI – indent 1
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 to be considered high-risk technologies;
Amendment 346 #
Motion for a resolution
Annex I – part A – point VI – indent 2
Annex I – part A – point VI – indent 2
- to monitor their compliance with thelaw and ethical principles set out in the proposed Regulation;
Amendment 349 #
Motion for a resolution
Annex I – part A – point VI – indent 3
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 351 #
Motion for a resolution
Annex I – part A – point VI – indent 4
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 ofa structured dialogue with all relevant stakeholders and civil society representatives.;
Amendment 354 #
Motion for a resolution
Annex I – part A – point VI – indent 4 a (new)
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 356 #
Motion for a resolution
Annex I – part A – point VI – indent 4 b (new)
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)
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)
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 362 #
Motion for a resolution
Annex I – part A – point VII
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 367 #
Motion for a resolution
Annex I – part B – recital 1
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 toshould 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 371 #
Motion for a resolution
Annex I – part B – recital 3
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 374 #
Motion for a resolution
Annex I – part B – recital 4
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 nationalcompetent supervisory authorityies is to be carried out in view of the objectives it is to pursue as laid down herein.
Amendment 376 #
Motion for a resolution
Annex I – part B – recital 5
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, deployed orment and used in the Union, including in cases where part of the technologies might be located outside the Union or not have a specificingle location, such as in the case of cloud computing services.
Amendment 380 #
Motion for a resolution
Annex I – part B – recital 6
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)
Annex I – part B – recital 6 a (new)
Amendment 382 #
Motion for a resolution
Annex I – part B – recital 6 b (new)
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 384 #
Motion for a resolution
Annex I – part B – recital 7
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 are 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 388 #
Motion for a resolution
Annex I – part B – recital 8
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 toshould 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 inexecution 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 394 #
Motion for a resolution
Annex I – part B – recital 9
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, environUnion’s legal principles and fundamental friendliness and sustainability, privacy and governance, should be considered high-risk from a compliance with ethical principles perspectiveghts should be considered high-risk where that is the conclusion of an impartial, regulated and external risk assessment by the national supervisory authorityCommission, 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 399 #
Motion for a resolution
Annex I – part B – recital 10
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 403 #
Motion for a resolution
Annex I – part B – recital 11
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 tor reverted to historicala previous state, identified as restoring safe functionalities in cases of non- compliance with those safety features.
Amendment 406 #
Motion for a resolution
Annex I – part B – recital 12
Annex I – part B – recital 12
(12) Developers, deployers and users arcan be responsible forto 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 safetythe abovementioned features, whereas deployers and users should deploy and use the concerned technologies in full observance of those features.
Amendment 408 #
Motion for a resolution
Annex I – part B – recital 13 a (new)
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 411 #
Motion for a resolution
Annex I – part B – recital 14
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
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 415 #
Motion for a resolution
Annex I – part B – recital 16
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 419 #
Motion for a resolution
Annex I – part B – recital 18
Annex I – part B – recital 18
(18) Software, algorithms and data used or producproduced 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 421 #
Motion for a resolution
Annex I – part B – recital 19
Annex I – part B – recital 19
(19) In line with Union law, software, algorithms and data used or producproduced 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
Annex I – part B – recital 20
Amendment 424 #
Motion for a resolution
Annex I – part B – recital 21
Annex I – part B – recital 21
(21) Artificial intelligence, robotics and related technologies, including software, algorithms and data used or producproduced or destined to be used by such technologies, should perform on the basis of sustainable progress. Such technologies should not run countribute comprehensively toer 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 426 #
Motion for a resolution
Annex I – part B – recital 22
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 producproduced or destined to be used by such technologies, should in no way causepurposefully 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 431 #
Motion for a resolution
Annex I – part B – recital 25
Annex I – part B – recital 25
(25) Socially responsible artificial intelligence, robotics and related technologies, including the software, algorithms and data used or producproduced 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
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 434 #
Motion for a resolution
Annex I – part B – recital 27
Annex I – part B – recital 27
Amendment 435 #
Motion for a resolution
Annex I – part B – recital 28
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 producproduced 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 an 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 444 #
Motion for a resolution
Annex I – part B – recital 31
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 reducending waste production, diminishing the carbon footprint, preventmedying climate change and avoiding environmental degradationpreserving 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
Annex I – part B – recital 33
(33) Any artificial intelligence, robotics and related technologies, including software, algorithms and data used or producproduced 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 448 #
Motion for a resolution
Annex I – part B – recital 34
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 biometricrecognition 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 454 #
Motion for a resolution
Annex I – part B – recital 36
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 459 #
Motion for a resolution
Annex I – part B – recital 37
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 461 #
Motion for a resolution
Annex I – part B – recital 38
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 nationalcompetent 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 466 #
Motion for a resolution
Annex I – part B – recital 39
Annex I – part B – recital 39
(39) Each nationalcompetent 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 468 #
Motion for a resolution
Annex I – part B – recital 40
Annex I – part B – recital 40
(40) NationalCompetent 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 470 #
Motion for a resolution
Annex I – part B – recital 41
Annex I – part B – recital 41
(41) NationalCompetent 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
Annex I – part B – recital 42
(42) Additionally, these nationalcompetent 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 482 #
Motion for a resolution
Annex I – part B – recital 46
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 nationalcompetent 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 489 #
Motion for a resolution
Annex I – part B – Article 2 – paragraph 1
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 producproduced or destined to be used by such technologies, developed, deployed or used in the Union.
Amendment 496 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point a
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 ordering, analysing, and interpreting their environment and taking actions, with some degree of autonomy, to reach conclusions or take acieve 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)
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
Annex I – part B – Article 4 – paragraph 1 – point b
(b) ‘robotics’ means technologies that enablautomatically controlled, reprogrammable, multi- purpose machines1a to perform tasks traditionally performed by human beingsactions in the physical world including by way of artificial intelligence or related technologies; __________________ 1aFrom the definition for industrial robots in ISO 8373.
Amendment 500 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point c
Annex I – part B – Article 4 – paragraph 1 – point c
Amendment 501 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point d
Annex I – part B – Article 4 – paragraph 1 – point d
Amendment 502 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point e
Annex I – part B – Article 4 – paragraph 1 – point e
Amendment 503 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point f
Annex I – part B – Article 4 – paragraph 1 – point f
Amendment 506 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point g
Annex I – part B – Article 4 – paragraph 1 – point g
Amendment 507 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point h
Annex I – part B – Article 4 – paragraph 1 – point h
Amendment 508 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point i
Annex I – part B – Article 4 – paragraph 1 – point i
Amendment 509 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point j
Annex I – part B – Article 4 – paragraph 1 – point j
Amendment 512 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point k
Annex I – part B – Article 4 – paragraph 1 – point k
Amendment 513 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point l
Annex I – part B – Article 4 – paragraph 1 – point l
Amendment 514 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point m
Annex I – part B – Article 4 – paragraph 1 – point m
Amendment 516 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point n
Annex I – part B – Article 4 – paragraph 1 – point n
Amendment 517 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point o
Annex I – part B – Article 4 – paragraph 1 – point o
Amendment 521 #
Motion for a resolution
Annex I – part B – Article 4 – paragraph 1 – point p
Annex I – part B – Article 4 – paragraph 1 – point p
Amendment 524 #
Motion for a resolution
Annex I – part B – Article 5 – paragraph 1
Annex I – part B – Article 5 – 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 accordance with Union law and fundamental rights, and the ethical principles laid down in this Regulation.
Amendment 536 #
Motion for a resolution
Annex I – part B – Article 6 – paragraph 2
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 539 #
Motion for a resolution
Annex I – part B – Article 6 – paragraph 3
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 542 #
Motion for a resolution
Annex I – part B – Article 7 – paragraph 1
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 Union law, fundamental rights, and the ethical principles set out in this Regulation shall be considered high-risk technologies.
Amendment 544 #
Motion for a resolution
Annex I – part B – Article 7 – paragraph 2
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 nationalcompetent supervisory authorities referred to in Article 14 or the European Agency for Artificial Intelligence.
Amendment 551 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point a
Annex I – part B – Article 8 – paragraph 1 – point a
Amendment 556 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point d
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 560 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point e
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 messageappropriately;
Amendment 561 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point f
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 564 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 1 – point h
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 historicala previous state, identified as restoring safe functionalities.
Amendment 567 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 2
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 nationalcompetent 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 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 3
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 nationalcompetent supervisory authorities referred to in Article 14 and other supervisory bodies.
Amendment 570 #
Motion for a resolution
Annex I – part B – Article 8 – paragraph 4
Annex I – part B – Article 8 – paragraph 4
Amendment 574 #
Motion for a resolution
Annex I – part B – Article 9 – paragraph 1
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 579 #
Motion for a resolution
Annex I – part B – Article 9 – paragraph 2
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 raceperceived 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 597 #
Motion for a resolution
Annex I – part B – Article 10 – paragraph 2 – point e
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 limitations and exceptions thereto;
Amendment 603 #
Motion for a resolution
Annex I – part B – Article 10 – paragraph 4
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 nationalcompetent 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 605 #
Motion for a resolution
Annex I – part B – Article 11 – paragraph 1
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 minimispreserves 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 609 #
Motion for a resolution
Annex I – part B – Article 11 – paragraph 2
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 addressremedy environmental issues such as waste production, the carbon footprint, climate change and environmental degradation.
Amendment 612 #
Motion for a resolution
Annex I – part B – Article 11 – paragraph 3
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 nationalcompetent 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 changeemergency and environmental degradation.
Amendment 615 #
Motion for a resolution
Annex I – part B – Article 12 – title
Annex I – part B – Article 12 – title
Privacy and biometric recognition
Amendment 618 #
Motion for a resolution
Annex I – part B – Article 12 – paragraph 2
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 Charterin 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 626 #
Motion for a resolution
Annex I – part B – Article 13 – paragraph 1
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 nationalcompetent supervisory authorities referred to in Article 14 and further supervisory bodies, in accordance with Union law, principles and values.
Amendment 627 #
Motion for a resolution
Annex I – part B – Article 13 – paragraph 2
Annex I – part B – Article 13 – paragraph 2
2. Data used or producproduced 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 633 #
Motion for a resolution
Annex I – part B – Article 14 – paragraph 1
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 nationalcompetent supervisory authority shall be responsible for assessing whether artificial intelligence, robotics and related technologies, including software, algorithms and data used or producproduced 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 637 #
Motion for a resolution
Annex I – part B – Article 14 – paragraph 2
Annex I – part B – Article 14 – paragraph 2
Amendment 638 #
Motion for a resolution
Annex I – part B – Article 14 – paragraph 2 a (new)
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 640 #
Motion for a resolution
Annex I – part B – Article 14 – paragraph 3
Annex I – part B – Article 14 – paragraph 3
3. Each nationalcompetent 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 stakeholdersand the cooperation and collaboration between stakeholders from academia, research, industry, civil society and individual experts .
Amendment 641 #
Motion for a resolution
Annex I – part B – Article 14 – paragraph 4
Annex I – part B – Article 14 – paragraph 4
4. Each nationalcompetent 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.