BRUSSELS — The European Union says its AI Act provides a framework to address risks posed by increasingly powerful artificial intelligence systems, as regulators continue to examine safety and compliance practices across the sector.
EU tech chief Henna Virkkunen said the legislation, adopted two years ago, includes risk-based requirements and oversight mechanisms designed to cover AI systems throughout their lifecycle. She rejected the suggestion that the rules are already outdated, arguing that future developments and potential risks had been considered.
The European Commission has requested safety and compliance information from more than 30 AI companies around the world, including firms in China. The responses are being assessed and could lead to further investigations where regulators identify possible breaches.
Under the AI Act, serious violations can carry fines of up to 7% of a company’s worldwide annual turnover, depending on the infringement. The Commission’s review comes amid broader international debate over how to govern advanced AI while balancing innovation, transparency and public safety.
Officials have also been considering a possible digital levy on large US technology companies as part of discussions on the EU’s 2028–2034 budget.
Illustrative image: server infrastructure.
