18 Amendments of Paul TANG related to 2020/2012(INL)
Amendment 22 #
Draft opinion
Paragraph 1
Paragraph 1
1. Believes that any ethical framework should seek to respect human autonomy, prevent harm, promote fairness, and respect the principle of explicability of technologies; the development, deployment and use of Artificial Intelligence (AI), and thus any ethical framework, need to fully respect fundamental rights, freedoms and values, including privacy, the protection of personal data, non-discrimination and the freedom of expression and information, as enshrined in the Union data protection law, namely Regulation (EU) 2016/679 of the European Parliament and of the Council (GDPR)1a and Directive (EC) 2002/58 of the European Parliament and of the Council (ePrivacy)1b, currently under revision, and the Charter of Fundamental Rights of the European Union; calls on the Commission to incorporate an obligation of transparency, explicability and the possibility of human intervention of AI applications in its ethical framework and to take care in its legislative proposals of sufficient oversight, enforcement with effective penalties, independent audits, specific stress tests and all other necessary means for a proper functioning checks and balances system; __________________ 1aRegulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ L 119, 4.5.2016, p. 1). 1bDirective 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications) (OJ L 201, 31.7.2002, p. 37).
Amendment 24 #
Draft opinion
Paragraph 1
Paragraph 1
1. Believes that any ethical framework shouldthere is a difference between ethics and law and the role they play in our societies; any framework of ethical principles for the development, deployment and use of Artificial Intelligence (AI), robotics and related technologies should complement the EU Charter of Fundamental Rights and thereby seek to respect human dignity and autonomy, prevent harm, promote fairness, and transparency, respect the principle of explicability of technologies; and guarantee that the technologies are there to serve people, with the ultimate aim of increasing human well-being for everybody;
Amendment 30 #
Draft opinion
Paragraph 1 – indent 1 (new)
Paragraph 1 – indent 1 (new)
- Independent audits on the application of AI should be conducted annually, in analogy with the financial sector, to examine whether the used AI- applications and checks and balances are in accordance with specified criteria; the auditors need to be supervised by an independent sufficiently overseeing authority;
Amendment 33 #
Draft opinion
Paragraph 1 a (new)
Paragraph 1 a (new)
1a. Recalls that AI may give rise to biases and thus to various forms of discrimination, such as sex, race, colour, ethnic or social origin, genetic features, language, religion or belief, political or any other opinion, membership of a national minority, property, birth, disability, age or sexual orientation; recalls that everyone’s rights must be ensured and that this principle of non- discrimination must be the core of the ethical framework of the Commission;
Amendment 39 #
Draft opinion
Paragraph 2
Paragraph 2
2. SHighlights the power asymmetry between those who employ AI technologies and those who interact and are subject to them; in this context stresses the importance of developing an “ethics-by-default and by design” framework which fully respect the Charter of Fundamental Rights of the European Union, Union law and the Treaties;
Amendment 44 #
Draft opinion
Paragraph 3
Paragraph 3
3. Considers that the current Union legalislative framework will need to be updaon protection of privacy and personal data fully applies to AI, robotics and related technologies, however could benefit from being supplemented with guidingrobust ethical principlguidelines; points out that, where it would be premature to adopt legal acts, a soft law framework should be used;
Amendment 66 #
Draft opinion
Paragraph 5 a (new)
Paragraph 5 a (new)
5a. Promotes a European Agency for Artificial Intelligence, which ensures a European coordination of AI standards and regulations; this centralized agency develops common criteria for a European certificate of ethical compliance, which also takes the data used for algorithmic processes into account;
Amendment 68 #
Draft opinion
Paragraph 5 b (new)
Paragraph 5 b (new)
5b. Promotes Corporate Digital Responsibility on a voluntary basis; the EU should support corporations, who by choice use digital technologies and AI ethically within their companies; the EU should encourage corporations to become proactive by establishing a platform for companies to share their experiences with ethical digitalization, as well as coordinating the actions and strategies of participating companies;
Amendment 76 #
Draft opinion
Paragraph 6
Paragraph 6
6. Stresses that the protection of networks of interconnected AI and robotics mustis important, and strong measures must be taken to prevent security breaches, cyber- attacks and the misuse of personal data;
Amendment 78 #
Draft opinion
Paragraph 6 a (new)
Paragraph 6 a (new)
6a. Calls for a comprehensive risk assessment of AI, robotics and related technologies in addition to the impact assessment provided by Article 35 GDPR (Article 27 of Directive (EU) 2016/680 and Article 39 of Regulation (EU) 2018/1725); the more impact an algorithm has, the more transparency, auditability, accountability and regulation is needed; where an algorithmic decision leads to a limitation of fundamental rights, there needs to be a very robust assessment in place; in highly critical fields - when health, freedom or human autonomy are directly endangered - the implementation of AI should be prohibited;
Amendment 91 #
Draft opinion
Paragraph 7
Paragraph 7
7. Notes that AI and robotic technology are used more and more in the area of law enforcement and border control could enhance public safety and security; stresses that its use must respect the principles of proportionality and necessity; , often with adverse effects on individuals when it comes to their rights to privacy, data protection and non- discrimination; stresses that the deployment and use of these technologies must respect the principles of proportionality and necessity, the Charter of Fundamental Rights, in particular the rights to data protection, privacy and non- discrimination, as well as the relevant secondary Union law such as EU data protection rules;
Amendment 98 #
Draft opinion
Paragraph 8
Paragraph 8
8. Stresses that AI and robotics are not immune from making mistakes and can easily have inherent bias; notes that biases can be inherent in the underlying datasets, especially when historical data is being used, introduced by the developers of the algorithms, or generated when the systems are implemented in the real world setting; considers the need for legislators to reflect upon the complex issue of liability in the context of criminal justice.
Amendment 106 #
Draft opinion
Paragraph 8 a (new)
Paragraph 8 a (new)
8a. Notes that AI is often used by digital platforms for automated decision making algorithms to disseminate and order the content shown to the users, including to organise their personal feed; stresses that these algorithms, how they work and how they order the shown material, are a black box to users, which takes away choice and control from the user, enables the creation of echo chambers and leads to a distrust in digital services; calls on the Commission to take the perspective of the end-user of AI- applications into account while drafting its ethical framework, with a focus on transparency, explicability and customization.
Amendment 110 #
Draft opinion
Paragraph 8 b (new)
Paragraph 8 b (new)
8b. Notes that AI can be used to analyse or predict aspects concerning individual's personal preferences, interests or behaviour, for profiling; emphasises that the quality of output of automated decision making AI is subject to the quality of used data and the chosen predetermined parameters; stresses that the use of automated decision making AI requires a strong legislative framework which protects privacy and personal data, and together with a duty of care obligation overseeing the legitimate use of the AI, which does not apply to content moderation, ensures full compliance; calls therefore on the Commission to work out a duty of care regime through detailed sectoral guidelines in order to use automated decision making algorithms in compliance with the fundamental rights of protection of personal data and privacy, laid down in the General Data Protection Regulation1a; __________________ 1aRecital 71 and article 22 General Data Protection Regulation.
Amendment 113 #
Draft opinion
Paragraph 8 c (new)
Paragraph 8 c (new)
8c. Points out that AI can be used for personalised advertising, in particular micro-targeted and behavioural advertisements, and of assessments of individuals; notes the potential negative impact of personalised advertisement, especially on minors, by interfering in the private life of individuals; posing questions as to the collection and use of the data used to personalise advertising, offering products or services or setting prices; perceives these downsides as expressions of the way personalised ads are unethical; calls therefore on the Commission to use its ethical framework to prohibit all personalised advertisements;
Amendment 114 #
Draft opinion
Paragraph 8 d (new)
Paragraph 8 d (new)
8d. Notes that AI can be used to manipulate face- and audiovisual characteristics, often referred to as deepfakes; recalls that this technique can be used to manipulate elections, to disseminate disinformation and for other undesirable actions; asks the Commission therefore to use its ethical framework to impose an obligation for all deepfake material or any other realistically made synthetic videos, to state it's not original and to introduce a strict limitation when used for electoral purposes;
Amendment 115 #
Draft opinion
Paragraph 8 e (new)
Paragraph 8 e (new)
8e. Emphasises the importance of the protection of personal data and privacy; observes the rapid development of AI applications to recognise unique characteristic elements, such as facial, movements and attitudes; warns for interferences of privacy, non- discrimination and the protection of personal data with the use of automated recognition applications; calls on the Commission to incorporate in its ethical framework an absolute ban on facial recognition in the public space and educational premises and a ban on not local storage of data used for facial recognition;
Amendment 116 #
Draft opinion
Paragraph 8 f (new)
Paragraph 8 f (new)
8f. Recalls the importance of linguistic and cultural diversity; calls therefore on the Commission to use its ethical framework to not let AI reduce this diversity, but to keep offer access to a wide variety of content which would not over- represent a single language and/or cultural model and to condemn any attempts from algorithms which would restrict this diversity and only offer content corresponding to some already existing patterns or which could act as an 'echo-chamber' that would prevent access to more diversity;