Aleph Alpha moderately welcomes the political compromise on the EU AI Act
12. December 2023, Heidelberg
Aleph Alpha has always advocated a moderate regulatory approach to artificial intelligence and, against this backdrop, welcomes the conclusion of the EU trilogue negotiations on the EU AI Act. Compared to some of the regulatory approaches originally discussed, a number of changes appear to have recently found their way into the regulations, which are likely to prevent overregulation in certain areas. However, many questions remain unanswered. It also remains important to see how AI regulation will develop in an international context. Regulation in Europe that is too strict in comparison remains a potential competitive disadvantage and a clear locational disadvantage.
The text of the political compromise reached on December 8, 2023 as part of the EU trilogue negotiations is not yet available. A final assessment of the individual regulations from the perspective of Aleph Alpha as a company that will be affected in the future is therefore not yet possible at this time.
However, it is already clear that numerous detailed questions remain unanswered due to a large number of undefined legal terms in the regulation – even after the regulation has been adopted by the EU legislator and entered into force. The resulting legal uncertainty must not be to the detriment of smaller AI providers such as Aleph Alpha. However, this risk remains.
On a positive note, the political compromise has succeeded in taking account of the rapid further development of AI technologies through dynamic regulatory approaches. For providers of AI models and AI systems, however, this increases the personal and technical effort involved in complying with regulatory requirements. The creation of a central EU authority, such as the AI Office, may be useful for coordinating open issues.
The emerging requirements for basic AI technologies (including so-called “general-purpose AI models”), but also for AI systems, are demanding. Their feasibility will also be determined by the course that is now being set in the technical implementation on the basis of the political compromise and the subsequent practical test. This will determine the extent to which the AI Act will restrict AI innovation from Europe and hinder the development of successful business models.
As Aleph Alpha, we have a clear ambition to implement the regulatory requirements in an exemplary manner when the AI Act comes into force. Until then, we would like to make a constructive contribution to clarifying open questions in the context of future regulation.
