Considering the fast development of AI technologies and the impact AI can have on individuals and the protection of their fundamental rights, the AI proposal of the European Commission (the AI Act) is an important initiative that is to be welcomed. However, the current proposal fails to provide adequate protection for citizens. The AI Act should ensure the protection of individuals against possible harm caused by AI technologies. It should provide procedural guarantees and access to legal remedies. Further, it should identify the actors who can be held accountable for potential damages and for the violation of rights.
In this comment, while considering the amendments in the Presidency compromise text of 29 November 2021, the Meijers Committee addresses specific gaps in protection in the AI proposal and provides some suggestions for improvement. Aside from the more detailed recommendations in the comments below, the Meijers Committee proposes to:
• Include individual safeguards to ensure transparency and accountability of the use of AI systems;
• Expand the scope of the AI Act to databases in the field of migration and asylum law;
• Prohibit the use of polygraphs, lie detectors, and social scoring systems.
22 February 2022