Guide
EU AI Act timeline
Updated June 16, 2026 · By Max Langley, AI Audits EU
The EU AI Act entered into force on 1 August 2024 and applies in phases. The bans and AI-literacy rules are already live. The big change for 2026 is that most high-risk obligations, originally due 2 August 2026, are being deferred to 2 December 2027 under the EU's Digital Omnibus. Here is the full sequence and what it means for planning.
The phased dates
- • 1 August 2024. The Act enters into force.
- • 2 February 2025. Prohibited AI practices and AI-literacy duties apply.
- • 2 August 2025. General-purpose AI model obligations apply.
- • 2 December 2027. Most high-risk obligations for standalone Annex III systems apply, deferred from the original 2 August 2026.
- • 2 August 2028. High-risk obligations for AI embedded in regulated products under Annex I apply.
Why the high-risk dates moved
The deferral comes from the Digital Omnibus on AI, a package the European Commission proposed to ease implementation while harmonised standards, notified-body capacity, and national authorities catch up. EU institutions reached a provisional political agreement in May 2026, confirmed by member states, and formal adoption is expected before August 2026. Until it is published in the Official Journal it is not final law, so treat 2 December 2027 as the planning date with that caveat in mind.
What this means for you
More time is not a reason to wait. The bans and GPAI rules already bite. Standing up a high-risk compliance programme takes months of documentation and assessment, and the standards you will rely on are still settling. Companies that prepare early also clear procurement faster, because EU buyers, partners, and insurers increasingly want documented conformity before they commit. The deadline moved; the work did not get smaller.
When does the EU AI Act take effect?
The Act entered into force on 1 August 2024 and applies in stages. Prohibited-practice and AI-literacy rules applied from 2 February 2025, general-purpose AI model rules from 2 August 2025, and most high-risk obligations were set for 2 August 2026. Under the EU's Digital Omnibus, those high-risk dates are being deferred.
Were the high-risk deadlines delayed?
Yes. The Digital Omnibus on AI defers most high-risk obligations for standalone Annex III systems from 2 August 2026 to 2 December 2027, and for high-risk AI embedded in regulated products under Annex I to 2 August 2028. The deferral was provisionally agreed by EU institutions in May 2026 and is expected to be formally adopted before August 2026.
What already applies today?
The bans on prohibited AI practices and the AI-literacy duties have applied since February 2025, and the general-purpose AI model obligations since August 2025. So parts of the Act are already enforceable even though the bulk of the high-risk regime is now scheduled for late 2027.
Should I wait until 2027 to prepare?
No. Building a risk management system, technical documentation, and a conformity assessment takes months, and harmonised standards and notified-body capacity are still maturing. Early preparation also wins procurement, since EU customers increasingly ask for documented conformity before they sign.
Sources
- Regulation (EU) 2024/1689 (the AI Act), Article 113 (application dates), EUR-Lex, eur-lex.europa.eu.
- European Parliament, Legislative train: Digital Omnibus on AI, europarl.europa.eu.
- European Commission, Regulatory framework on artificial intelligence, digital-strategy.ec.europa.eu.
Use the extra time well
A readiness assessment now means you are documented and procurement-ready long before the 2027 deadline arrives. Tell us what you run.
This guide is general information, not legal advice. Confirm your obligations with qualified counsel.