European Union lawmakers agreed to postpone the AI Act's core obligations for high-risk AI systems. The Council and Parliament reached a provisional deal on May 7, 2026, and Parliament voted to adopt it on June 16. Rules for stand-alone high-risk systems, such as AI used in hiring and employment, now apply from December 2, 2027 instead of August 2, 2026. High-risk AI built into regulated products such as medical devices has until August 2, 2028.
The agreement also delays labeling requirements for AI-generated content to December 2, 2026 for systems already on the market, extends some small-business exemptions to small mid-cap companies, and keeps a simplified registration requirement for some systems. A new ban on AI that generates child sexual abuse material or non-consensual intimate images applies from December 2, 2026.
For U.S. businesses, this matters mainly if you sell into the EU, have EU employees, or buy software from vendors that operate there. The delay buys time but does not remove the obligations. If you use AI tools for hiring or HR decisions that touch EU residents, ask the vendor how they plan to meet the 2027 requirements and what documentation they will provide. Keep an inventory of where AI is used in your business, which is useful under any regulation.
The Council still had to give formal approval, after which the text is published in the EU's Official Journal and takes effect three days later.
Sources
Does this affect your business?
Our advisors can help you figure out the right move, at no cost.