Europe achieved a provisional deal on Friday regarding landmark European Union rules governing the use of artificial intelligence (AI). The deal covers a range of areas, including governments’ use of AI in biometric surveillance and the regulation of AI systems like ChatGPT.
Following nearly 15 hours of negotiations, EU countries and European Parliament members reached a political agreement, marking a significant step towards Europe becoming the first major world power to enact laws specifically governing AI. The deal was reached after an intensive 24-hour debate the previous day.
European Commissioner Thierry Breton stated at a press conference, “Europe has positioned itself as a pioneer, understanding the importance of its role as a global standard setter. This is yes, I believe, a historical day.” The agreement sets transparency obligations for foundation models such as ChatGPT and general-purpose AI systems (GPAI) before they are introduced to the market.
High-impact foundation models will undergo model evaluations, assess and mitigate systemic risks, conduct adversarial testing, and report to the European Commission on serious incidents. Governments’ use of real-time biometric surveillance in public spaces is restricted to cases of specific crimes, prevention of threats, and searches for individuals suspected of serious crimes.
The agreement explicitly bans cognitive behavioral manipulation, untargeted scraping of facial images, social scoring, and biometric categorization systems inferring personal beliefs. Consumers are granted the right to launch complaints, with fines for violations ranging from 7.5 million euros ($8.1 million) or 1.5% of turnover to 35 million euros or 7% of global turnover.
While praised for its ambitious approach, the rules faced criticism from business group DigitalEurope, citing additional burdens for companies. Privacy rights group European Digital Rights expressed concerns about the legalization of live public facial recognition.
The legislation is expected to enter into force early next year, applying two years after formal ratification. With governments worldwide seeking to balance the advantages of AI with regulatory safeguards, Europe’s comprehensive AI rules may set a precedent for other nations.
