Source: IEU Monitoring

Brussels, 7 May 2026: European co-legislators reached a political agreement on the Digital Omnibus on AI to simplify implementation of the AI Act while preserving protections for society, safety and fundamental rights. The deal fixes staggered application dates: obligations for AI systems used in certain high-risk areas (including biometrics, critical infrastructure, education, employment, migration, asylum and border control) will apply from 2 December 2027, while rules for AI integrated into products such as lifts or toys will apply from 2 August 2028, allowing time for standards and support tools to be prepared. Watermarking obligations for AI-generated content have been postponed to 2 December 2026.

The agreement also strengthens citizen protection by prohibiting AI systems that generate non-consensual sexually explicit or intimate content and child sexual abuse material, explicitly targeting AI “nudification” apps; the prohibition covers placing such systems on the EU market for that purpose, placing them without reasonable safety measures, and deployers who use them to create such content, with images, video or audio included. Other changes include extending certain SME privileges to small mid-cap companies, clarifying that machinery products must comply with sectoral safety rules rather than duplicate AI requirements, narrowing the definition of “safety component,” allowing limited processing of personal data to detect and correct biases with safeguards, expanding access to regulatory sandboxes including an EU-level sandbox, and strengthening the Commission AI Office’s enforcement powers for oversight of certain general-purpose models and systems embedded in very large online platforms and search engines. The provisional deal must be formally adopted by Parliament and Council, after which amendments will be published in the Official Journal and enter into force three days later.