Article 113 — Application calendar

EU AI Act enforcement timeline

The EU AI Act stages its application across six milestones, governed by Article 113. Entry into force was on 1 August 2024; the substantive obligations apply on staggered dates between February 2025 and (following the Digital Omnibus deferral) August 2028. The Annex III high-risk obligations once keyed to 2 August 2026 are now provisionally deferred to 2 December 2027. Below is what enforces on each date and which AiExponent article spoke covers it in depth.

Source: Regulation (EU) 2024/1689 Article 113 — Official Journal.

2024-08-01

Live

Entry into force

Regulation (EU) 2024/1689 entered into force on the twentieth day after publication in the Official Journal (Art. 113 first sentence). Most operative provisions apply later under the staggered calendar below.

2025-02-02

Live

Application

Chapter I (general provisions including Art. 4 AI literacy) and Chapter II (prohibited AI practices, Art. 5) become applicable. Per Art. 113(a).

See the article spoke →

2025-08-02

Live

Application

Chapter V (general-purpose AI models — Articles 51–56), Chapter III Section 4 (notifying authorities and notified bodies), Chapter VII (governance and AI Office), most of Chapter XII (penalties — Articles 99 and 100), and Art. 78 confidentiality. Per Art. 113(b). Important exception: Article 101 (Commission-imposed GPAI fines) is explicitly excluded by Art. 113(b) — those fines do not become applicable until 2 August 2026 under the default Art. 113 date.

See the article spoke →

2026-08-02

Upcoming

Application

Article 101 — Commission-imposed GPAI fines (excluded from the 2025-08-02 date by Art. 113(b)) become applicable today, the default Art. 113 application date. Note: the high-risk system obligations (Articles 8–15, 16–27, Article 19) that originally also keyed to this date have been DEFERRED to 2 December 2027 under the Digital Omnibus (see next milestone) — that deferral is provisional and pending adoption in the Official Journal.

See the article spoke →

2027-12-02

Upcoming

Application

Annex III high-risk system obligations apply — Articles 8–15 (risk management, data governance, technical documentation, record-keeping, transparency, human oversight, accuracy and robustness), Articles 16–27 (provider, importer, distributor, deployer obligations), and Article 19 log retention. DEFERRED from the original 2 August 2026 under the Digital Omnibus. This date is PROVISIONAL — based on the Council–Parliament political agreement of 7 May 2026 and not yet adopted or published in the EU Official Journal.

See the article spoke →

2028-08-02

Upcoming

Application

Article 6(1) and Annex I — high-risk AI systems that are safety components of, or themselves are, products covered by sector-specific Union harmonisation legislation listed in Annex I (medical devices, machinery, in vitro diagnostics, civil aviation, marine equipment, toys, etc.). DEFERRED from the original 2 August 2027 under the Digital Omnibus (provisional, pending adoption in the Official Journal). Also: GPAI transitional deadline — providers of general-purpose AI models placed on the EU market before 2 August 2025 must come into full compliance with Article 53 obligations by 2 August 2027 (Art. 111(3), unchanged).

When the calendar pressure becomes a programme

Tools surface deadlines. Programmes meet them.

The dates above are non-negotiable. The work between today and each milestone — programme design, board narrative, regulator dialogue — is the surface AskAjay covers, the advisory arm of AI Exponent LLC.

Explore advisory at AskAjay.ai →