Europe Prepares Rules for Frontier AI Systems
The European Union is set to gain new powers to police large general purpose AI models, with the Commission able to demand information from model providers, conduct safety evaluations, and impose fines of up to 3% of annual total turnover from August 2.
These models power chatbots, image generators, and enterprise tools, and the new mandate will place the EU at the global frontier of AI safety regulation.
The AI Act enters its enforcement era after years of negotiation and regulatory fine-tuning, but its success depends on the Commission’s willingness to use its new powers.
Strong enforcement would signal that Brussels is ready to uphold its AI rules, even if it creates friction with the US or China, home to most leading model providers, including those that may be urged to adopt certain models for better regulation.
Risto Uuk, head of European policy and research at the Future of Life Institute, said, “Enforcement depends on the willingness at a political level to enforce the law and take it seriously.”
The rules target providers of the largest general purpose AI models, such as OpenAI, Anthropic, Meta, Alphabet, xAI, and China’s Alibaba, ByteDance, and Z.ai, which are expected to fall within scope.
France’s Mistral is the only European model provider that should fall under the law, and obligations under the AI Act for these models are governed by Article 55, which requires companies to perform state-of-the-art model evaluations and assess and mitigate systemic risk.
Related: Trump tariffs divide American consumers
They have built infrastructure to enforce the rules, including publishing the Code of Practice, issuing guidelines for model providers, and establishing an AI Act Advisory Forum.
The AI Office’s first enforcements are unlikely to involve blockbuster fines.
Rather, they will likely involve requests for information, structured dialogue with model providers, and independent evaluations.
It will depend on avoiding politicization, prioritizing the largest safety risks, and being adaptable to rapid technological changes.
Europe enters this new phase with more international support than it has enjoyed in the past, with states in the US advancing their own frontier AI rules.
Risto Uuk noted, “Europe should not feel entirely alone in this,” with the geopolitical tide turning towards greater regulation of these frontier models.
