47 Amendments of Andreas SCHIEDER related to 2020/2012(INL)
Amendment 2 #
Draft opinion
Paragraph 1
Paragraph 1
1. Highlights that the security and defence policies of the European Union and its Member States are guided by the principles enshrined in the European Charter of Fundamental Rights and those of the United Nations Charter, and by a common understanding of the universal values of the inviolable and inalienable rights of the human person, of freedom, of democracy, of equality and of the rule of law; highlights that all defence- related efforts within the Union framework must respect these universal values whilst promoting peace, security and progress in Europe and in the world;
Amendment 5 #
Draft opinion
Paragraph 1
Paragraph 1
1. Underlines the importance of an EU regulatory framework being applicable where consumers within the Union are users of or subject to an algorithmic system, irrespective of the place of establishment of the entities that develop, sell or employ the system; furthermore, believes that the rules set out should apply across the value chain, namely development, deployment and use of the relevant technologies and their components and for all developers, and should guarantee the highest level of consumer protection; proposes that these rules take into account the lessons drawn from the adaptation to Regulation (EU) 2016/6791a (GDPR), considered a global benchmark; considers the designation of a liable entity in the Union (such as authorised representative) important for its enforcement; __________________ 1aRegulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1).
Amendment 7 #
Draft opinion
Paragraph 2
Paragraph 2
2. Stresses that a Union framework regulating the use of artificial intelligence (AI)-enabled systems in defence must respect all applicable legal regimes, in particular the international humanitarian law and the international human rights law, and be in compliance with Union law, principles and values; stresses that the EU should play a global role in leading the way towards a credible and binding AI regulatory agenda rooted in democratic values; calls on the Union to assess the inherent AI-related risks with regard to the application of Union law, and foresee necessary adjustment and enforcement where needed; underlines that emerging technologies not covered by international law should be judged by the principle of humanity and the dictates of public conscience; underlines that the ethics of AI-enabled systems in defence must be assessed from the point of view of Human rights, and notably human safety, health and security, freedom, privacy, integrity and dignity;
Amendment 9 #
Draft opinion
Paragraph 2
Paragraph 2
2. Stresses that a Union framework regulating the use of artificial intelligence (AI)-enabled systems in defence must respect all applicable legal regimes, in particular the international humanitarian law and the international human rights law, and be in compliance with Union law, principles and values; the framework must therefore indicate the likeliness of errors and inaccuracies to deployers for the deployment of AI technology; calls on the Union to assess the inherent AI-related risks with regard to the application of Union law, and foresee necessary adjustment and enforcement where needed;
Amendment 9 #
Draft opinion
Paragraph 2
Paragraph 2
2. Notes that the framework should apply to algorithmic systems, including the fields of artificial intelligence (AI), internet of things (IoT), machine learning, deep learning, automated decision making processes and robotics; further notes that referral systems should be developed to help explain those systems to consumers whenever they present complexity or constitute decisions that impact their lives significantly;
Amendment 14 #
Draft opinion
Paragraph 2 a (new)
Paragraph 2 a (new)
2a. Emphasises that the geographical scope of such a framework should cover all the components of artificial intelligence, robotics and related technologies developed, deployed or used in the Union, including in cases where part of the technologies might be located outside the Union or not have a specific location;
Amendment 15 #
Draft opinion
Paragraph 2 a (new)
Paragraph 2 a (new)
2a. Underlines the need of constant monitoring of the use of AI; especially from the point of view of its advantages and disadvantages, as well as, its impact on the protection of Universal Human Rights;
Amendment 16 #
Draft opinion
Paragraph 2 b (new)
Paragraph 2 b (new)
2b. Recalls that the impact of AI systems should be considered not only from an individual perspective but also from the perspective of a society as a whole; calls to fully incorporate in a new framework, the human-centric approach based on the Communication on Building Trust in Human-Centric AI and the input obtained in the Ethics Guidelines prepared by the High-Level Expert Group on AI;
Amendment 21 #
Draft opinion
Paragraph 3
Paragraph 3
3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm or breaches of rights for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations and certification requirements should gradually increase with the identified risk level; in the lowest risk category there should be no additional legal obligations; algorithmic systems that may harm an individual and certification or labelling systems should be voluntary; algorithmic systems that may harm an individual or cause potential breaches of an individual’s rights, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk-based approach should follow clear and transparent rules, providing enough legal certainty whilst being future-proof;
Amendment 24 #
Draft opinion
Paragraph 3 a (new)
Paragraph 3 a (new)
3a. Calls for establishing of synergies and networks between the various European research centres on AI as well as other multilateral fora, such as: Council of Europe, the United Nations Educational Scientific and Cultural Organization (UNESCO), the Organisation for Economic Co-operation and Development’s (OECD),the World Trade Organisation and the International Telecommunications Union (ITU), in order to align their efforts and to better coordinate the developments of the AI technology;
Amendment 26 #
Draft opinion
Paragraph 3
Paragraph 3
3. Stresses that any future regulation should follow a differentiated risk-based approach, based on the potential harm for the individual as well as for society at large, taking into account the specific use context of the algorithmic system; legal obligations should gradually increase with the identified risk level; in the lowest risk category there should be no additional legala labelling obligations; algorithmic systems that may harm an individual, impact an individual’s access to resources, or concern their participation in society shall not be deemed to be in the lowest risk category; this risk- based approach should follow clear and transparent rules;
Amendment 30 #
Draft opinion
Paragraph 3 a (new)
Paragraph 3 a (new)
3a. Believes that the ethical principles should be the basis for a harmonised European system of risk classification and related legal obligations;
Amendment 31 #
Draft opinion
Paragraph 4
Paragraph 4
4. Underlines the importance of an ethical and regulatory framework, including in particular provisions on the quality of data sets used in algorithmic systems in relation to the purpose of its use, especially regarding the representativeness of training data used, on the de-biasing of data sets, as well as on the algorithms themselves, and on data and aggregation standards; stresses that those data sets should be auditable and made available to the competent authorities whenever called upon to ensure their fitness with the previously exposed principles;
Amendment 36 #
Draft opinion
Paragraph 4
Paragraph 4
4. Highlights that, based on a human- centric approach, the Unionwhereby technology fully respects Human Rights and humans retain authority over automated decision- making systems, the Union needs a robust AI regulatory framework focused on security and defence, followsing a path of responsibility, of protecting our citizens, and of defending our values, that its policies aim at preserving peace, preventing conflicts and strengthening international security, whilst seizing the opportunities that those technologies offer;
Amendment 43 #
Draft opinion
Paragraph 4 a (new)
Paragraph 4 a (new)
4a. Calls on the Member States and the European Commission to ensure that the algorithms used in defence systems, while keeping the necessary confidentiality, are governed by the principle of transparency, including a clear liability regime for the results of AI use; underlines that these algorithms must be constantly adjusted to the progress in AI technologies;
Amendment 46 #
Draft opinion
Paragraph 5
Paragraph 5
5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, process, rationale, reasoning and possible outcome of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected; recalls that humans must always be able to overrule automated decisions that are final and permanent;
Amendment 47 #
Draft opinion
Paragraph 5
Paragraph 5
5. Underlines that the Union must be at the forefront of mastering those technologies by establishing well defined processes for their use, for understanding the related ethical aspects and for fostering an effective international regulatory framework that contains the inherent risks of these technologies and prevents use for malicious purposes; the Union working together with the Member States must determine the appropriate liability regimes applicable to innovations in AI and other immersive technologies in the field of security and defence thus establishing a legal basis for accountability and traceability mechanisms, those include in particular unintended harm to persons, be it material or immaterial, such as breach of fundamental rights;
Amendment 47 #
Draft opinion
Paragraph 5
Paragraph 5
5. Believes that consumers should be adequately informed in a timely, impartial, easily-readable, standardised and accessible manner about the existence, the aims or purpose, process, rationale, reasoning and possible outcome and consequences for consumers of algorithmic systems, about how to reach a human with decision- making powers, and about how the system’s decisions can be checked, meaningfully contested and corrected;
Amendment 57 #
Draft opinion
Paragraph 6
Paragraph 6
6. Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible trustworthiness standards as regards the collection and exploitation of operational datareliability standards, active monitoring and supervision as regards the collection and exploitation of operational data; AI systems and applications intended to extract and synthesise data, and extrapolate results therefrom to inform decisions for matters relating to defence and national security, must be specific in scope and comply with the provisions set out in the current regulatory framework in terms of gathering and processing data; stresses that AI applications designed to process data for intelligence purposes in defence related activities should comply with data processing standards to avoid risks of unintended surveillance or infringement of individual rights; believes that for high-risk applications of AI-enabled technologies like facial recognition which lack a definitive regulatory framework at the EU level, the Union must ensure that their development and deployment is rightful, proportional and respects the rights of individuals; stresses the importance of monitoring competent national law enforcement authorities which develop and deploy AI-enabled systems and technologies to maintain public order so as to mitigate the disproportional risks of predictive policing;
Amendment 59 #
Draft opinion
Paragraph 6
Paragraph 6
6. Stresses that for any defence application of AI enabled systems, the Union should set technical and organisational standards to ensure their resilience against cyber-attacks and digital influence, as well as their compliance with the highest possible trustworthiness standards as regards the collection and exploitation of operational data; draws attention to the need of careful analysis of the algorithms on which AI makes its decisions; emphasises the importance for transparency and accountability of AI algorithms; notes the important distinction between transparency of algorithms and transparency of the use of algorithms;
Amendment 63 #
Draft opinion
Paragraph 7
Paragraph 7
7. Stresses that where public money contributesmoney originating from public sources contributes significantly to the development or implementation of an algorithmic system, the code, the generated data - as far as it is non-personal - and the trained model should be public by default, to enable transparency and reuse, among other goals, to maximise the achievement of the Single Market, and to avoid market fragmentation;
Amendment 65 #
Draft opinion
Paragraph 6 a (new)
Paragraph 6 a (new)
6a. Notices the great risk of AI in the area of disinformation; underlines that, if not regulated, AI technologies might have ethically adverse effects by exploiting bias in data and algorithms that may lead to disinformation, creating information bubbles and exploiting biases incorporated into AI algorithms;
Amendment 68 #
Draft opinion
Paragraph 7
Paragraph 7
7. Highlights the need to adopt clear safety and security provisions and requirements, with proper certifications, for AI-systems in defence, and carry ourt constant monitoring, regular tests and verifications across the entire life cycle; underlines the necessity of ensuring compliance with applicable standards and obtained certifications where AI modifies e.g. through machine learning the functionality and behaviour of systems in which it is integrated, in order to ensure full traceability of decisions made with involvement of AI;
Amendment 69 #
Draft opinion
Paragraph 7 a (new)
Paragraph 7 a (new)
7a. Underlines that the increased use of artificial intelligence requires a strong focus on digital security, as the large amount of data creates new risks of cyberattacks; calls on the Commission to develop clear guidelines for businesses and public agencies to take the necessary precautions when using artificial intelligence;
Amendment 71 #
Draft opinion
Paragraph 7 a (new)
Paragraph 7 a (new)
7a. Calls for a uniform implementation of the system of risk classification and related legal obligations to ensure a level-playing field among the Member States and to prevent a fragmentation of the internal market;
Amendment 72 #
Draft opinion
Paragraph 7 a (new)
Paragraph 7 a (new)
7a. Highlights the fact that AI framework in defence and security should develop benchmarks for ethically responsible and accepted uses of AI technologies; underlines that these criteria must be constantly adjusted to the progress in AI technologies;
Amendment 75 #
Draft opinion
Paragraph 7 b (new)
Paragraph 7 b (new)
7b. Considers that AI, IoT, and other emerging technologies have enormous potential to deliver opportunities for consumers to have access to several amenities in many aspects of their lives alongside with better products and services, as well as to benefit from better market surveillance, as long as all applicable principles, conditions (including transparency and auditability), and regulations continue to apply;
Amendment 76 #
Draft opinion
Paragraph 7 b (new)
Paragraph 7 b (new)
7b. Calls for the establishment of an Union-wide registration system for artificial intelligence, robotics and related technologies to support the uniform and transparent implementation of the risk classification in the Union;
Amendment 81 #
Draft opinion
Paragraph 8
Paragraph 8
8. Stresseds that all AI-systems in defence must have a concrete and well- defined domain of use and must be endowed withwhereby humans retain the abilitgency to detect and, disengage or diseactivate deployed systems should they move from their domain of use or engage in any escalatory or unintended action;
Amendment 81 #
Draft opinion
Paragraph 8
Paragraph 8
8. Underlines the importance of ensuring that the interests of consumers and groups which are marginalised andor in vulnerable consumers and groupsituations are adequately taken into account and represented in any future regulatory framework; notes that for the purpose of analysing the impacts of algorithmic systems on consumers, access to data should be extended to appropriate parties notably independent researchers, media and civil society organisations, while fully respecting Union data protection and privacy law; recalls the importance of training and giving basic skills to consumers to deal with algorithmic systems in order to protect them from potential risks and detriment of their rights;
Amendment 85 #
Draft opinion
Paragraph 9
Paragraph 9
9. Underlines that the entire responsibility for the decision to design, develop, deploy and use AI-systems must rest on human operators and, as there must be meaningful human control over any weapons system and human intent in the decision to use force; underlines that the human-in- the-loop principle must also be applied to the command and control of AI- enabled systems; stresses that AI-enabled systems must allow the military leadership to assume its full responsibility and exercise the necessary level of judgmentaccountability for the use of lethal force and exercise the necessary level of judgment, which cannot be endowed to machines as it must be based on distinction, proportionality and precaution, for taking lethal or large-scale destructive action bey means of such systems; recalls in this respect its position on a ban on the development, production and use of fully autonomous weapons systems enabling strikes to be carried out without meaningful human intervention;
Amendment 87 #
Draft opinion
Paragraph 8 a (new)
Paragraph 8 a (new)
8a. Considers that, when it comes to filling the gaps on the practical implementation of ethical guidelines on algorithmic systems and connected technologies, models such as the so-called VCIO model (Values, Criteria, Indicators, Observables) should be studied and evaluated in terms of fitness for purpose; further considers that such models to concretise and implement AI system requirements, as well as to produce labels that allow companies to communicate the ethical properties of their products clearly and uniformly through a standardised risk matrix, could enhance consumer literacy, information, and awareness;
Amendment 94 #
Draft opinion
Paragraph 9 a (new)
Paragraph 9 a (new)
9a. Stresses that the EU must take the lead in promoting the establishment of international norms regarding the ethical and legal parameters of the development and use of fully autonomous, semi- autonomous and remotely operated lethal weapons systems; Member States should develop national strategies for the definition, status and use of lethal autonomous weapons (LAWs) towards a comprehensive strategy on the EU level;
Amendment 99 #
Draft opinion
Paragraph 10
Paragraph 10
10. Calls for the Union to establishment of a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities;n the framework of a European Agency for Artificial Intelligence, which the Parliament proposed in its resolution of 16 February 20171a, issuing guidance, opinions and expertise to Member States’ authorities, monitoring the implementation of relevant EU legislation, addressing potential consumer protection issues, identifying standards for best practice, and, where appropriate, making recommendations for regulatory measures; __________________ 1aEuropean Parliament resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics (2015/2103(INL)) (OJ C 252, 18.7.2018, p. 239).
Amendment 100 #
Draft opinion
Paragraph 10
Paragraph 10
10. Calls for the Union to consider the establishment of a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities; further calls for this structure to be appropriately advised by stakeholder organisations (such as consumer protection organisations) in order to ensure wide consumer representation;
Amendment 102 #
Draft opinion
Paragraph 10
Paragraph 10
10. Calls for the Union to establish a European market surveillance structure for algorithmic systems issuing guidance, opinions and expertise to Member States’ authorities; emphasizes that Member States must develop risk-management strategies for AI in the context of their national market surveillance strategies.
Amendment 103 #
Draft opinion
Paragraph 10
Paragraph 10
10. Underlines that the Union must promote understanding of the military implications of AI, of robotics and of autonomy; considers that the Union needs to promote the acquisition of the necessary skills and knowledge on technology development processes and operational methods throughout the supply chain and over the full lifecycle of AI-enabled military capabilities;
Amendment 108 #
Draft opinion
Paragraph 10 a (new)
Paragraph 10 a (new)
10a. Underlines, that despite its added value, AI comes with a number of weaknesses, one of them relating to all sorts of different types of biases; emphasises that AI technologies should clearly be void of any sort of profiling, especially regarding gender;
Amendment 108 #
Draft opinion
Paragraph 11
Paragraph 11
11. Notes that it is essential for the software documentation, the algorithms and data sets used or produced by artificial intelligence, robotics, and related technologies to be fully accessible to market surveillance authorities, while respecting Union law; further notes that such elements should be preserved by those who are involved in the different stages of the development of algorithmic systems and in proportion of their liability, namely through distributed ledger technologies, such as block-chain; invites the Commission to assess if additional prerogatives should be given to market surveillance authorities in this respect;
Amendment 116 #
Draft opinion
Paragraph 11
Paragraph 11
11. Stresses the need to overcome the current fragmentation within the Union as regards national AI-related law, research, innovation and expertise in the area of AI, which puts in jeopardyendangers the functioning of the internal market and the objective to ensure trustworthyreliable and secure development of AI in Europe; in this respect welcomes the inclusion of AI- related projects under the European Industrial Development Programme (EDIDP); believes that the future European Defence Fund (EDF) and the Permanent structured cooperation (PESCO) also offer well adapted frameworks for future AI- related projects that would help to better streamline Union efforts in this field; stresses that AI-related projects should be synchronized with the wider EU civilian programmes devoted to AI; notes that in line with the European Commission’s White Paper on AI (COM2020/65final) excellence and testing centres concentrating on research and development of AI in the field of security and defence should be established with vigorous specifications underpinning the participation of and investment from private stakeholders;
Amendment 116 #
Draft opinion
Paragraph 12
Paragraph 12
12. Calls for the designation by each Member State of a competent national authorityenforcement body(NEB) for monitoring the application of the provisions;
Amendment 121 #
Draft opinion
Paragraph 11 a (new)
Paragraph 11 a (new)
11a. Emphasises that the development of AI that respects fundamental rights and supports the public interest requires the strategic pooling and sharing of data in the EU between private and public entities, as well as the strengthening of an EU AI ecosystem, which involves public, private, and civil society stakeholders; calls on the European Commission to foster dialogue among Member States, researchers, academics, civil society actors and the private sector so as to have inclusive policymaking processes when it comes to defence-related AI regulations;
Amendment 125 #
Draft opinion
Paragraph 13
Paragraph 13
13. Calls for the establishment of a European market surveillance board for algorithmic systems in the framework of a European Agency for Artificial Intelligence, to ensure a level playing field and to avoid fragmentation of the internal market, to decide with a qualified majority and by secret vote in case of different decisions on algorithmic systems used in more than one Member State, as well as at the request of the majority of the national authorities;
Amendment 126 #
Draft opinion
Paragraph 11 a (new)
Paragraph 11 a (new)
11a. Calls on the Commission to present their "Reinforcement of the Skills Agenda", announced in the White Paper on Artificial Intelligence on the 19th February2020, as soon as possible - in order to ensure that everyone in Europe can benefit from the digital transformations of the EU economy;
Amendment 128 #
Draft opinion
Paragraph 13 a (new)
Paragraph 13 a (new)
13a. Believes that a European certification of ethical compliance should be designed in such a way as to inform consumers about the risk level of a product or a service with an algorithmic component as well as its trustworthiness in the light of the ethical principles and all other requirements based on relevant Union legislation;
Amendment 129 #
Draft opinion
Paragraph 13 a (new)
Paragraph 13 a (new)
13a. Stresses the need for national market surveillance authorities to be reinforced in terms of capacity, skills, and competences in AI, IoT and other related technologies, as well as knowledge about its specific risks;
Amendment 133 #
Draft opinion
Paragraph 12
Paragraph 12
12. Calls on the European Commission and on the VP/HR to present, also as part of an overall approach, a sectoral AI strategy for defence-related activities within the Union framework, that ensures both respect for citizens’ rights and EU’s strategic interests that should propose a consistent regulatory approach spanning from the inception of AI-enabled systems to their military uses; calls on these regulatory efforts to be supported by meaningful monitoring schemes, so that normative frameworks are not outplaced by technological development and new methods of warfare; calls on the Council, the European Commission and on the VP/HR to enter in a structured dialogue with the European Parliament to that end.