The French Secretary of State for the Digital Transition, Cédric O, presented the priorities of the French Presidency of the Council of the EU (FPEU) in the digital field to the members of the Special Committee on Artificial Intelligence (AI) in the Digital Age (AIDA) on Thursday 10 February (see EUROPE 12884/10).
“The European Union is caught in a dilemma regarding technology, between two contradictory injunctions. The first is the technology gap. Of the top 10 companies in the world, eight are in the technology sector. Six did not exist 25 years ago and none are European. And the second injunction is that of regulation in the face of the fears and consequences of artificial intelligence in particular, which is today an element of democratic unease”, the French Secretary of State began.
In view of this, Cédric O confirmed that the efforts of the FPEU would be focused on making progress in the discussions by the end of June, while deeming the Commission’s proposal “balanced”.
Furthermore, Mr O also considered that one of the main challenges in developing an AI framework in the EU will be “to maintain a single market”.
In order to achieve this, the approach should be prominently based on an analysis of the existing Regulations, their strengths and weaknesses. “We need harmonisation to allow for a unified Regulation”, insisted Cédric O.
Furthermore, the French Secretary of State said, the focus should also be on limiting the risk of “crowding out” AI services developed by European companies.
“If we consider that a European company must have legitimate constraints to develop AI algorithms, we must guarantee that a system developed under different conditions in the United States or in Asia will not be able to access the European market”, he said.
Regulatory sandboxes and fundamental rights
Following Cédric O’s speech, several members of the European Parliament’s AIDA Committee expressed their concerns, notably about ‘regulatory sandboxes’, which allow, for example, actors in a sector to test the service developed without being obliged to comply with the entire regulatory framework for a limited period of time.
The French Secretary of State conceded that this process would be “indispensable” and that the applicable rules should also be different for small start-ups and large companies.
However, Mr O was more discreet when asked by Spanish MEP Iban García del Blanco (S&D) about the need - or not - for an EU-wide coordination body for AI.
“The EU Council does not yet have a position, but it is possible to draw parallels with some other texts such as the Digital Services Act (DSA) (see EUROPE 12885/7). Perhaps we could take inspiration from some of the things we have done, but in saying this I am already going outside the competence as Presidency of the EU Council”, he said.
Finally, some MEPs also insisted on not losing sight of the societal issues behind the development of AI systems. “Businesses need certainty and we also need to make choices to guarantee fundamental rights”, commented Kim van Sparrentak (Greens/EFA, Netherlands). (Original version in French by Thomas Mangin)