Members of the European Parliament’s Committee on Civil Liberties (LIBE) and the Committee on the Internal Market (IMCO) debated, on Tuesday 25 January, the European Commission’s proposal for harmonised rules in the field of artificial intelligence (AI) (see EUROPE 12869/13).
Described by the European Commission in the introduction to the debate as a potential future “international model” the text was generally welcomed by MEPs. However, some took the opportunity to point out the major challenges and pitfalls to be avoided.
Among the main concerns about the future development of AI is the issue of mass surveillance. “Some practices are prohibited, but there are exceptions, such as real-time biometric identification. We cannot risk that too broad an interpretation by Member States will lead to abuses”, said Brando Benifei (S&D, Italy).
A view shared by other MEPs, such as Dragoș Tudorache (Renew Europe, Romania), who is concerned that the distinction between legal and repressive uses is difficult.
In addition, the debate also focused on the definition of AI itself. Some, like Deirdre Clune (EPP, Ireland), advocate an approach that would specify areas not covered by the text.
In addition, several MEPs questioned the effectiveness of self-certification processes for the marketing of certain AI systems.
“This needs to be explored further. After completing this self-certification, the problem is not solved, some products could be declared compliant without actually being compliant”, said Brando Benifei.
Finally, the issues of innovation and regulatory overload also came up several times during the discussions.
“We have an important responsibility towards companies, to keep them in Europe and not to let them leave because of too much administrative and regulatory burden”, summarised Svenja Hahn (Renew Europe, Germany). (Original version in French by Thomas Mangin)