European Union Officially Publishes Landmark AI Act With Staggered Deadlines
The EU officially publishes its AI Act in the Official Journal, triggering a phased rollout that begins with bans on certain AI systems in early 2025. Full compliance requirements for high-risk systems will not take effect until 2026 and 2027.
The European Union officially publishes the AI Act in the Official Journal, marking a major milestone in global artificial intelligence regulation. This final text closely mirrors previous drafts and establishes harmonized rules for AI across the bloc. The regulation enters into force twenty days after publication, but its rules do not take effect all at once.
Instead, the AI Act follows a staggered implementation schedule based on the type of AI system involved. Prohibitions on certain unacceptable AI practices and new AI literacy requirements take effect on February 2, 2025. Rules for general-purpose AI models follow on August 2, 2025, while most obligations for high-risk systems and transparency requirements apply starting August 2, 2026.
A final batch of requirements for specific high-risk systems listed in Annex I will not become applicable until August 2, 2027. The law also includes exemptions or extended deadlines for certain AI models and systems that are already on the market. This gradual approach gives businesses time to adjust their operations and come into full compliance.