In May this site reported that the EU was about to delay the AI Act’s high-risk rules. It has now done so. Regulation (EU) 2026/1744, known as the Digital Omnibus on AI, was published in the Official Journal on 24 July 2026 and entered into force three days later.
The dates that moved
2 December 2027. Obligations for stand-alone high-risk AI systems, the Annex III list that includes AI used in employment, education and access to essential services. The original date was 2 August 2026.
2 August 2028. Obligations for high-risk AI built into products that are already regulated, such as medical devices and machinery.
The dates that did not move
February 2025. The prohibited practices.
August 2025. The obligations on providers of general-purpose AI models.
2 August 2026. The transparency duties: telling people when they are dealing with an AI system and labelling deepfakes. Systems already on the market before that date have until 2 December 2026 to meet the requirement to mark AI-generated content.
What else changed
The Omnibus added a prohibition on AI systems that generate non-consensual intimate images or child sexual abuse material. It extended the simplified documentation rules to medium-sized companies. It widened the AI Office’s supervisory role over general-purpose models.
What this means
The delay buys time for the hardest part of compliance. It does not remove anything. An organisation using AI in hiring, lending or education now has fourteen months to classify its systems, fix its contracts with suppliers and build human oversight that works.
The first deadline is not 2027. For anything that generates content, it is 2 December 2026.
This article is general information on the law as at November 2026. It is not legal advice and has not been updated since that date. See the terms of use.