Digital Omnibus political agreement: May 2026 update
The Commission announced a political agreement on 7 May 2026. It is a major signal for high-risk AI timing, but teams should still distinguish it from formally adopted law.
What the official announcement says
The Commission says the political agreement sets a clearer implementation timeline for high-risk AI systems. Rules for certain high-risk areas, including biometrics, critical infrastructure, education, employment, migration, asylum and border control, would apply from 2 December 2027. High-risk AI embedded in products such as lifts or toys would apply from 2 August 2028.
What has not changed as current law yet
The AI Act Service Desk still presents 2 August 2026 as the current-law date for the majority of rules, including Annex III high-risk rules and Article 50 transparency obligations. Until the amending text is formally adopted and published, internal planning should keep the current-law baseline visible.
- Update planning decks with a clearly labelled “political agreement” column.
- Do not pause AI literacy, GPAI, Article 50 design, inventory, or evidence collection.
- For high-risk systems, keep classification and evidence work moving, but avoid over-locking final technical files before the amended timeline and support tools are settled.
FAQ
Is the AI Act officially delayed now?
Not as current law. The political agreement is a strong signal, but formal adoption and publication still matter before treating the new dates as binding.
Which proposed date matters most for Annex III high-risk systems?
The Commission announcement points to 2 December 2027 for many high-risk areas, with 2 August 2028 for high-risk AI embedded in regulated products.
Turn this reading into an actionable report
Use the free scanner to map your likely role, detect likely obligations, and see which evidence is missing.