Activities of Manuel BOMPARD related to 2020/2012(INL)
Shadow opinions (1)
OPINION with recommendations to the Commission on a framework of ethical aspects of artificial intelligence, robotics and related technologies
Amendments (40)
Amendment 2 #
Draft opinion
Recital A
Recital A
A. Whereas the Union is founded on the ethical values stated in Article 2 of the Treaty on European Union and on compliance with the precautionary principle stated in Article 191(2) of the Treaty on the Functioning of the European Union;
Amendment 7 #
Draft opinion
Recital B
Recital B
B. Whereas the artificial intelligence (AI) global leadership race, which may well bring about a race to the bottom as regards national regulations; whereas, therefore, the European Union should grasp this opportunity to map out a regulatory framework that will determine the source of ethical values and standards shaping the sector worldwide, is picking up the pace;
Amendment 14 #
Draft opinion
Recital C
Recital C
C. Whereas AI solutions may benefit society in the areas of green transition, environment protection, waste management, climate change, energy management and efficiency, air quality e.g. smart grids and electro-mobility; whereas, however, it is essential to minimise the ecological footprint of robotics, as the use of cyber-physical systems and robots is expected to increase overall energy consumption and the amount of electrical and electronic waste; whereas, lastly, 14% of current jobs in the OECD area could disappear over the next 15 to 20 years because of automation, with 32% undergoing radical change;
Amendment 19 #
Draft opinion
Recital C a (new)
Recital C a (new)
Ca. Whereas the number of those who will be durably affected by the deployment of AI is huge; whereas, in this connection, Microsoft journalists were recently laid off and replaced by AI carrying out the same content-sorting task; whereas, in addition, AI has regularly pushed articles from questionable information sources and indeed fake news;
Amendment 22 #
Draft opinion
Recital C b (new)
Recital C b (new)
Cb. having regard to AI energy performance; whereas a standard machine-learning project now produces some 284 tonnes of carbon dioxide equivalent throughout its development cycle, which is five times more than the emissions from a car from when it is manufactured until it is scrapped; whereas simultaneously pushing the Internet of Things and action to combat the climate crisis is therefore absurd and societies must bring about major change in production and consumption;
Amendment 25 #
Draft opinion
Recital C c (new)
Recital C c (new)
Cc. Whereas guideline 6 in the Commission communication 'Building Trust in Human-Centric Artificial Intelligence', based on the work of the group of experts, focuses on sustainability and ecological responsibility of AI systems; whereas acknowledgement of the finite nature of Earth's resources is an ethical stance in terms of Union policy, as demonstrated by the Green Deal or the declaration of a state of climate and environmental emergency; whereas the AI ethical framework must therefore encompass ecodesign and component recyclability criteria;
Amendment 27 #
Draft opinion
Recital D
Recital D
D. Whereas AI can be applied to almost any field in medicine: biomedical research, exemplified by the AI-discovered antibiotic Halicin or AI contributions to new cancer therapies, medical education, clinical decision-making, personalizsed medicine, psychiatric diagnosis and treatment, in revolutionizsing robotic prostheses and support systems, telemedicine and the overall efficiency of the health systems; whereas this requires appropriate training and preparation for health and administrative personnel; whereas digital health should not be a gateway to dehumanised care; whereas, in addition, there is a considerable risk that the digital divide, if not resolved, will be instrumental in the spread of 'medical deserts' across Europe;
Amendment 31 #
Draft opinion
Recital D a (new)
Recital D a (new)
Da. Whereas personal health data must remain a patient's property; whereas no information concerning the health of patients must be passed on without their full and informed consent; whereas personal health data, which by their nature are extremely sensitive, must be scrupulously safeguarded; whereas the highest cybersecurity standards ought to be established for the relevant networks; having regard in particular to the interesting initiatives emerging from a number of AI developers during the crisis that offer hospitals anti-spam solutions to safeguard their e-mail systems free of change, and subject to no obligation, against cyberattacks;
Amendment 38 #
Draft opinion
Recital E
Recital E
E. Whereas current policy and ethical guidelines for AI are lagging behind ethical challenges that must be identified and mitigated, since AI has tremendous capability to threaten patient preference, safety, and privacy; whereas the boundaries between the roles of physicians and machines in patient care need to be outlined in accordance with the principle of supervised robot autonomy, under which the initial programming of care and the final decision as to how it is to provide it, including diagnostics, remain within the decision-taking sphere of the appropriate physician or care personnel;
Amendment 42 #
Draft opinion
Recital E a (new)
Recital E a (new)
Ea. Whereas Union data protection rules should be adapted to take into account the increasing complexity and interconnectivity of care and medical robots that may handle highly sensitive personal information and health data and should be consistent with privacy by design as established by Regulation (EU) 2016/679 on data protection;
Amendment 44 #
Draft opinion
Recital F
Recital F
F. Whereas AI solutions may benefit society in the area of food safety and Farming 2.0,data aggregation in connection with digital farming systems is handled by means of data platforms held by large business groups; whereas, accordingly, the digital tools used within farming tend to focus on mass production and disregard small and medium-sized producers; whereas the Union holds leadership in AI applicationshigh-tech solutions encouraged within farming encourage more intensive and standardised production, thus strengthening the global biodiversity trend;
Amendment 49 #
Draft opinion
Recital F a (new)
Recital F a (new)
Fa. Whereas, both globally and at European level, AI is one of the 21st century's strategic technologies; whereas in January 2017 the Chinese electrical- appliance business Midea completed acquisition of the European firm KUKA AG, one of the world's leading robotics, installation technology and intelligent- systems engineering suppliers; whereas that takeover should have been a wake-up call with regard to protecting Europe's AI sector, in particular SMEs and start-ups; whereas EU Competition Commissioner Margrethe Vestager has alerted Member States to the risks surrounding Chinese corporate acquisitions during the coronavirus crisis; whereas the Commission has pointed out that, in the context of the COVID-19 crisis, there could be a greater risk of attempts to acquire healthcare capacity (e.g. to produce medical devices or protective equipment) or facilities carrying out related activities such as research institutes (e.g. to develop vaccines) through foreign direct investment;
Amendment 53 #
Draft opinion
Recital F b (new)
Recital F b (new)
Fb. Whereas exponential growth in the development of AI and its incorporation into applications, products and services has been conducive to the emergence, in particular via the platform economy, of new business and service models; whereas work will be reorganised as a result of AI, producing a state of hyperconnectivity accompanied by more intensive work, increased stress and working time and a blurring of work-life boundaries; whereas this development has major consequences for workers' mental and physical health because it is a potential factor triggering occupational illnesses such as burnout or particular forms of social anxiety;
Amendment 54 #
Draft opinion
Recital F c (new)
Recital F c (new)
Fc. Whereas the development of AI is conducive to the emergence of microtaskers working on on-line platforms whose tasks are digitally apportioned among them by algorithms; whereas platform microtasking is growing exponentially; whereas in 2013, according to the World Bank, there were 145 on-line work platforms, attracting 50 million crowdworkers; whereas those figures are now likely to be out of date; whereas platform microtasking is mainly carried out by vulnerable and precarious groups such as retired persons, students, unemployed people, migrants and people with disabilities; whereas these workers are constantly confronted with algorithmic reputation scores and automated tests and are constantly at risk of being disconnected from platforms, without redress, in the event of poor scores over an extended period; whereas they are therefore particularly exposed to precariousness, social isolation, burnout, and work and a pace of life that have no structure, with major consequences for their physical and mental well-being;
Amendment 55 #
Draft opinion
Recital F d (new)
Recital F d (new)
Fd. Whereas robotic machines blur the boundaries between human subjects and technological objects; whereas not only do they have implications for society that need to be ethically assessed, but they also challenge the very ethical frameworks on the basis of which they are to be assessed; whereas, as is pointed out in the report by the World Commission on the Ethics of Scientific Knowledge and Technology (COMEST), particular attention should be paid to the use of medical robots, nursing robots, care robots for the elderly and companion robots;
Amendment 56 #
Draft opinion
Recital F e (new)
Recital F e (new)
Fe. Whereas the use of social robots and companion robots is spreading rapidly within healthcare and, in particular, within elderly care; whereas care robots for the elderly and companion robots may take on a functional and emotional role; whereas those robots may have a role to play in reducing loneliness among older people, preventing behaviours associated with dementia, stimulating the cognitive activities of patients with a neurodegenerative disease or performing particular everyday tasks that are difficult for elderly persons to carry out; whereas companion robots may thus provoke feelings that are false, illusory and unreciprocated, deluding and infantilising older people;
Amendment 57 #
Draft opinion
Recital F f (new)
Recital F f (new)
Ff. Whereas companion robots may increasingly be used for sexual purposes; whereas the use of sex robots that look like children or are programmed to be abused has particularly worrying ethical implications;
Amendment 62 #
Draft opinion
Paragraph 1 a (new)
Paragraph 1 a (new)
1a. Takes note of the first definition of AI proposed by the Commission in its April 2018 communication on artificial intelligence for Europe; notes the need to lay down a common definition of the concepts of robot and artificial intelligence that is technology-neutral and sufficiently flexible in order to ensure the legal certainty conducive to stimulating essential investment in this area in the EU and to ensuring public trust and consumer protection;
Amendment 70 #
Draft opinion
Paragraph 2
Paragraph 2
2. Underlines that overregulation may hampere fact that legal safeguards may stimulate AI sector innovation, especially for SMEs and Sstart- ups; considers that hindering the Union AI sector in delivering benefits of AI applications in e.g. healthcare, environment protection and food quality to the citizens, may also bear ethical weight, especially in the context of global competition, where securing full respect of Union ethical values may pose a challengepoints out, especially in the context of global competition, that, with the GDPR, the European Union has led the way towards a binding ethically rigorous model;
Amendment 74 #
Draft opinion
Paragraph 2 a (new)
Paragraph 2 a (new)
2a. Points out that the OECD's ethical framework takes account of labour market upheaval; stresses that automation combined with AI will increase productivity and therefore increase output; points out that, as during previous technological revolutions, some jobs will be replaced; stresses that increased use of robotics and AI should also reduce human exposure to harmful and hazardous conditions and should also help to create more quality and decent jobs and improve productivity; points to the work of the OECD, which stresses that automation may give society the option to cut the number of hours worked, thus improving workers' living conditions and health;
Amendment 78 #
Draft opinion
Paragraph 2 b (new)
Paragraph 2 b (new)
2b. Stresses that social dialogue must be used to ensure a fair transition for workers as AI is deployed and tasks become automated, and this transition should include access to training programmes throughout their working lives, support for those affected by job cuts and access to new opportunities on the labour market; stresses that firms should also invest in training and reskilling their existing workforce with a view to addressing their needs;
Amendment 80 #
Draft opinion
Paragraph 2 c (new)
Paragraph 2 c (new)
2c. Draws further attention to the OECD recommendations calling for governments to work closely with stakeholders to promote the responsible use of AI at work, to enhance the safety of workers and the quality of jobs, and to aim to ensure that the benefits of AI are broadly and fairly shared;
Amendment 81 #
Draft opinion
Paragraph 3
Paragraph 3
3. Welcomes the fact that the Risk- Based Approach methodology defined in the Commission White Paper of 19 February 20201 recognises healthcare, transport and energy as high risk sectors by default, and proposes to introduce listed AI requirements beyond existing-Unionhorizontal rules inand these sectors, unless the manner in which AI is used does not involve significant risk; stresses that the Union AI ethical framework should address especially the above high-risk introduction of a list of more stringent AI requirements beyond existing Union rules in these sectors; _________________ 1White Paper On Artificial Intelligence - A European approach to excellence and trust, COM(2020)0065
Amendment 84 #
Draft opinion
Paragraph 3 a (new)
Paragraph 3 a (new)
3a. Stresses that the development and use of AI, including deep learning methods, has a significant environmental impact because it increases overall energy consumption, the volume of data stored, and the amount of electric and electronic waste; proposes that two measures be introduced in the new Circular Economy Action Plan – the European waste- reduction and resource-efficiency strategy: the setting of an overall target of halving the EU's material footprint by 2030 and of a ceiling for absolute reduction of waste per capita, by carrying out direct action to prevent waste generation and introducing a specific sectoral target for some waste, such as electronic waste;
Amendment 87 #
Draft opinion
Paragraph 3 b (new)
Paragraph 3 b (new)
3b. Highly recommends that the Member States and the Commission promote the use of due diligence in the trade of metals needed to build electronic and robotics technologies;
Amendment 88 #
Draft opinion
Paragraph 3 c (new)
Paragraph 3 c (new)
3c. Insists that European green data centres should be promoted and developed with a view to reducing our dependency on foreign and private data centres, subject to European rules and standards, with a medium-term objective of 100% renewable energy use;
Amendment 90 #
Draft opinion
Paragraph 3 d (new)
Paragraph 3 d (new)
3d. Stresses the scale of the ecological footprint of AI and robotics, as discussed in guideline 6 of the Commission communication 'Building Trust in Human-Centric Artificial Intelligence'; insists that the environmental impact of these areas needs to be limited through energy-efficient consumption, by promoting the use of renewable technologies for robotics and the use and re-use of secondary raw materials, and by cutting down on electric and electronic waste; encourages the Commission, therefore, to incorporate circular economy principles into all the EU's robotics policies;
Amendment 97 #
Draft opinion
Paragraph 4
Paragraph 4
4. Notes that, with the rapid development of AI and the uncertainty that lies ahead, a common Union AI ethical framework will expand an ecosystem of trust, including for users, as defined in the Commission White Paper, whetherparticularly in environment protection, healthcare or food safety applications, thus supporting the ecosystem of excellence in legal certainty and providing effective response to the challenges yet not defined in courtrooms, management meetings or scientific laboratories;
Amendment 112 #
Draft opinion
Paragraph 5 a (new)
Paragraph 5 a (new)
5a. Stresses that the health sector has specific characteristics; agrees with the Commission that citizens should have secure access to a comprehensive electronic record of their health data, and should remain in control of and be able to share their personal health data securely with authorised parties, while, in accordance with the data protection legislation, unauthorised access should be prohibited; notes, further, that data should be stored on secure local servers and processed by independent bodies;
Amendment 117 #
Draft opinion
Paragraph 5 b (new)
Paragraph 5 b (new)
5b. Stresses that neither insurance companies nor any other type of service provider should be authorised to use e- health data to introduce discrimination in the setting of prices, given that this would run counter to the fundamental right to the highest attainable standard of health;
Amendment 119 #
Draft opinion
Paragraph 5 c (new)
Paragraph 5 c (new)
5c. Points out that, to take decisions, robots use algorithms which play the part of values and ethical frameworks, and that their introduction has significant ethical implications for healthcare and social relations; is particularly concerned about the use for paedophilic and sexual abuse purposes of companion robots; believes that ethical considerations should be taken into account in the design of robotics technologies; calls, in the development process for these machines, for a place to be granted to ethics, based on an approach such as value-sensitive design, particularly with regard to care robots for the elderly and companion robots; stresses that this approach should also be adjusted to take account of animal welfare;
Amendment 122 #
Draft opinion
Paragraph 6
Paragraph 6
6. Calls for securing the full enforcement of Union legal framework on data protection and privacy, relevant notably in the healthcare AI applications and related sensitive data, to strengthen the “Right to an explanation” foreseen in Article 22 of Regulation (EU) 2016/679 of the European Parliament and of the Council (General Data Protection Regulation, (GDPR))2 and higher interpretability requirements for high-risk AI; stresses, on the basis of Article 7 of this regulation, the need for freely given and informed consent; calls for the establishment of a duty of explainability for decisions taken by AI within the ethical framework; _________________ 2Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).
Amendment 129 #
Draft opinion
Paragraph 6 a (new)
Paragraph 6 a (new)
6a. Stresses the connection between reflection on the AI ethical framework and the European data strategy; calls for the introduction of a 'data of public interest' or 'data of general interest' concept for private data that could be opened up for the purposes of a public research goal (energy production and consumption, biodiversity data, climate data, etc.), in line with work by the Internet Governance Forum;
Amendment 134 #
Draft opinion
Paragraph 6 b (new)
Paragraph 6 b (new)
6b. Stresses that algorithms and machines that have the capacity to learn and evolve should be designed in a transparent and respectful manner from the outset, with particular attention given to their implications for workers' and consumers' physical and mental well- being; stresses that workers should be able to benefit from a right to an explanation on decisions taken by algorithms with a view to reducing uncertainty and opacity, which are harmful to workers' well-being in the long term; calls on the Commission and Member States to legislate to that effect; urges the Commission to put forward a legislative initiative regarding platform micro-taskers to provide them with legal protections, which are essential to their physical and mental well-being;
Amendment 140 #
Draft opinion
Paragraph 7
Paragraph 7
7. Supports the view that the seven AI requirements identified in the Ethics Guidelines for Trustworthy AI of the High- Level Expert Group on AI constitute solid building blocks for a common Union AI ethical framework, addressing, among others, ethical aspects of AI applications in environment, health and food protection; calls for an improvement of the acquis on transparency, traceability and human oversight, which were indicated as areas in need of further improvement in the feedback given on the Guidelines by 350 organisations; furthermore, encourages the creation of the Union AI ethical framework in a spirit of openness to the works of other international partners that share Union values, e.g. to the Rome Call for AI Ethics by Pope Francisparticularly the UN and the Council of Europe, which has drawn up a European ethical charter on the use of artificial intelligence in judicial systems and has its own legal research centre, the Ad Hoc Committee on Artificial Intelligence (CAHAI);
Amendment 145 #
Draft opinion
Paragraph 8
Paragraph 8
8. Highlights the benefits of AI for disease prevention and control, exemplified by AI predict and climate change monitoring, particularly ing the case of zoonoses, such as COVID-19 epidemic before WHO; urges the Commission to equip ECDC in its reform, which was brought about by, whose development and spread is exacerbated by deforestation; stresses that the development and use of AI tracking technologies to respond to the COVID-19 crisis, with the legal framework and resources allowing for gathering necessary data independently, including, among others, AI solutionpandemic has raised a number of ethical questions on users' rights; stresses, further, that some tracing applications were clearly violating the EU's ethical values;
Amendment 152 #
Draft opinion
Paragraph 8 a (new)
Paragraph 8 a (new)
8a. Calls for the EU not to slip into 'technology solutionism' and to maintain a human role in decisions and public policies; considers that, particularly in the health sector, it is useful for robots to be present to support the work of doctors or healthcare assistants, with a view to improving the human experience of diagnosis and treatment, without, however, disregarding the need to ensure that medical practice and patient care practices are not dehumanised;
Amendment 164 #
Draft opinion
Paragraph 9 a (new)
Paragraph 9 a (new)
9a. Criticises the fact that the development of digital farming promised by the Commission as part of the CAP reforms consolidates the stranglehold of an industrial farming system that is very harmful for the environment and biodiversity; calls for digital technologies to serve to develop a European farming model based on agro-ecology; calls for preference to be given to open-source platforms, run directly by farmers, providing mutual assistance and sharing good practices among farmers, and putting producers of foodstuffs in direct contact with consumers through short supply chains; insists that farmers must keep ownership and control over data produced on their holdings, even after aggregation;
Amendment 168 #
Draft opinion
Paragraph 9 b (new)
Paragraph 9 b (new)
9b. Calls on the Commission to establish a temporary ban on buying European AI and robotics industry businesses that are currently undervalued or facing commercial troubles as a result of the coronavirus crisis;
Amendment 170 #
Draft opinion
Paragraph 9 c (new)
Paragraph 9 c (new)
9c. Supports the Commission in the establishment of a list of assets and businesses in 'critical' domains, such as health, medical research, biotechnology, digital, artificial intelligence, robotics and infrastructure essential to our security and to public order; calls for a ceiling to be set to prevent foreign businesses in critical domains from acquiring majority stakes;