EU Seats AI Act Expert Panels as High-Risk Rules Slip to Late 2027

The European Commission has appointed a Scientific Panel of 60 independent experts and a new Advisory Forum to support enforcement of the Artificial Intelligence Act, putting the bloc's AI enforcement machinery in place just as lawmakers prepare to push the law's toughest compliance deadlines back by as much as 16 months.

Announced on 1 June 2026, the appointments capture where European AI regulation stands in mid-2026. The Commission is building the structures to police the world's first comprehensive AI law while a package of amendments agreed on 7 May 2026, the Digital Omnibus on AI, delays obligations for high-risk systems from 2 August 2026 to 2 December 2027 and writes a ban on AI "nudifier" tools into the statute.

Both new bodies will advise the Commission's AI Office and national authorities on applying the rules. Members serve two-year terms.

"The Scientific Panel brings together 60 world-leading independent experts with experience in frontier AI, engineering, technical auditing, industry and societal impact."

European Commission press release, 1 June 2026

Its brief centers on general-purpose AI models and systems, taking in systemic risks, model classification, evaluation methodologies and cross-border market surveillance, according to the Commission.

The Advisory Forum is broader. Its members come from academia, civil society and industry, including small and medium-sized enterprises and startups, and it will advise on standardization and implementation challenges. The EU Agency for Fundamental Rights and the cybersecurity agency ENISA hold permanent seats, alongside European standardization bodies.

The experts arrive while the rules they will help enforce are still moving. On 7 May 2026, negotiators from the Council of the European Union, the European Parliament and the Commission reached a provisional agreement on the Digital Omnibus on AI, the first set of amendments to the AI Act since its adoption in June 2024.

The headline change is timing. Obligations for high-risk systems listed in Annex III of the Act, which covers uses such as biometrics, critical infrastructure, education, employment, law enforcement and border management, slip from 2 August 2026 to 2 December 2027. For AI embedded in regulated products like medical devices, lifts and radio equipment, the deadline moves a full year, to 2 August 2028.

Smaller deferrals follow the same pattern. Providers of systems that generate synthetic content and were already on the EU market before 2 August 2026 now have until 2 December 2026 to make their outputs detectable as artificially generated. And member states get until 2 August 2027 to set up national AI regulatory sandboxes.

Why the slippage? The compliance tooling is not ready. Law firm Covington & Burling noted in a 2 June analysis that the delays give European standards bodies such as CEN-CENELEC additional time to prepare the necessary standards and guidelines that will serve as the backbone for several requirements.

The omnibus is not all relief. It adds two new prohibited practices to Article 5 of the Act: AI systems that generate or manipulate non-consensual intimate imagery of identifiable people, and systems that produce child sexual abuse material. The ban takes effect on 2 December 2026. Under the agreed text, providers escape liability only where they have implemented reasonable and adequate technical safety measures to reliably prevent such output.

Parliament had campaigned for the nudifier ban since late March, when it adopted its negotiating position by 569 votes to 45, with 23 abstentions.

Fines got a touch sharper too. Breaches of the Act's newly tightened information-sharing duties across the AI supply chain now sit in the first-tier penalty band: up to 3% of worldwide annual turnover or €15 million.

None of it is law yet. The amendments need formal sign-off from Parliament and the Council, and Covington expects final approval in June with publication in the EU's Official Journal in July; the changes would enter into force on the third day after publication. The schedule is tight by design, since the package has to be law before 2 August 2026, the date the current high-risk rules would otherwise begin to apply. The new expert panels start their two-year terms with the rulebook they are meant to interpret still in motion.