The Federal Association of the Digital Economy (BVDW) e. V. participated with a statement in the consultation of the EU Commission on the proposal for a regulation for a European concept for artificial intelligence (AI). The proposed regulation includes bans on certain practices in the field of AI, special requirements for high-risk AI systems and obligations for operators of such systems, harmonized transparency regulations for AI systems and regulations for market observation and market surveillance. The BVDW generally advocates a Europe-wide approach to dealing with artificial intelligence, but at the same time raises the question of whether such a uniform regulation, as presented by the Commission, is the appropriate means to make AI meaningful and profitable in the European Union.
Artificial intelligence is a location factor for the European economy
“From the BVDW’s point of view, trustworthy and innovative artificial intelligence is an essential location factor for Europe and its economy. Compliance with European values and principles is of the highest relevance and people – as well as the effects on them – should be the focus of all efforts, ”explains Achim Himmelreich from Capgemini, who is Vice President of the BVDW.
From the BVDW’s point of view, comprehensive basic regulation appears premature against the background of ongoing scientific and economic developments. In addition, it is questionable whether such a regulation can be implemented at all and whether legal certainty can be created as a result. The goal of creating the first comprehensive basic regulation for AI thus appears to be very ambitious.
Artificial intelligence harbors both risks and opportunities
“From the BVDW’s point of view, the details of the regulation reveal that a horizontal approach that applies to all AI applications, with the variety of functions and areas of application of AI, entails massive difficulties and legal uncertainties,” says Himmelreich.
Because AI is used in a wide variety of areas and brings with it different challenges in the applications. And accordingly it also harbors various risks and opportunities – also with a view to questions of data protection law. If AI is used in the health sector, for example, numerous unresolved legal issues could be resolved using a sectoral approach. The BVDW therefore calls on the Commission to choose a sectoral approach instead of a horizontal approach and to make proportionate adjustments.
AI regulation as part of the legal framework for data
“The AI regulation must also be part of a coherent legal framework for data that ensures that AI – even if under certain requirements to protect valuable goods – is not disproportionately restricted, for example through parallel legislative proposals,” explains Himmelreich.
From the BVDW’s point of view, the overly broad definition of AI and existing legal uncertainties are particularly problematic. Also with regard to high-risk AI, contradictions in the transparency obligations, the sometimes disproportionate requirements and the bureaucracy. The latter are associated with high fines, among other things.
Also read: Digital economy: New members in the BVDW strengthen digital offeringste