Statistics
What are the EU AI Act deadlines?
The EU AI Act deadlines run from 2 February 2025 to 31 December 2030. The high-risk dates were amended in July 2026 and now start in December 2027.
By the Foxnut team · Updated
What are the EU AI Act deadlines?
The EU AI Act deadlines run in stages from 2 February 2025 to 31 December 2030. The Act entered into force on 1 August 2024 under Article 113. The high-risk rules now apply from 2 December 2027 and 2 August 2028, because the Digital Omnibus on AI amended those dates with effect from 27 July 2026.
We are Foxnut AI, a senior-only practice in Bengaluru and Paris that builds AI systems for companies. We are not your lawyers, and this is not legal advice. There’s a link to talk to us at the end of this article. Let’s get you the correct dates first.
What does this EU AI Act timeline record?
This timeline records every application date of the EU AI Act, Regulation (EU) 2024/1689, as the dates stand today, with a primary source for each row.
It exists because the EU AI Act deadlines you find elsewhere are frequently the original ones. The amendment that moved the high-risk dates entered into force on 27 July 2026, and summaries written before it are stale on exactly the rows that matter. Each date below carries what applies, who it reaches, and the article or regulation it comes from.
12 Jul 2024
The AI Act, Regulation (EU) 2024/1689, is published in the Official Journal of the European Union
1 Aug 2024
The Act enters into force, on the twentieth day after publication, per Article 113
2 Feb 2025
Chapters I and II apply: the prohibited AI practices of Article 5 and the AI literacy obligation of Article 4
2 Aug 2025
Obligations for general-purpose AI model providers (Chapter V), governance (Chapter VII), notified bodies (Chapter III Section 4), penalties except Article 101 (Chapter XII) and confidentiality (Article 78) apply
2 Aug 2026
The Act's general application date: everything not given its own date, including the Article 50 transparency obligations; Article 101, under which the Commission can fine general-purpose AI model providers up to 3% of worldwide turnover or EUR 15 million; and Article 57's requirement that every member state have an operational AI regulatory sandbox
2 Aug 2027
General-purpose AI models placed on the market before 2 August 2025 must be brought into compliance
2 Dec 2027
Rules for high-risk AI systems in the Act's listed areas - including biometrics, critical infrastructure, education, employment, migration, asylum and border control - apply, as amended by the Digital Omnibus on AI
2 Aug 2028
Rules for high-risk AI embedded in regulated products - such as lifts or toys, where AI is a safety component under EU product law - apply, as amended by the Digital Omnibus on AI
2 Aug 2030
High-risk AI systems intended for use by public authorities that were already on the market before the rules applied must be brought into compliance
31 Dec 2030
AI components of the EU's large-scale IT systems (the Act's Annex X) placed on the market before 2 August 2027 must be brought into compliance
| Number | What it measures | Period | Source |
|---|---|---|---|
| 12 Jul 2024 | The AI Act, Regulation (EU) 2024/1689, is published in the Official Journal of the European Union | 12 July 2024 | EUR-Lex, Regulation (EU) 2024/1689, linked in the sources below |
| 1 Aug 2024 | The Act enters into force, on the twentieth day after publication, per Article 113 | 1 August 2024 | EUR-Lex, Regulation (EU) 2024/1689, Article 113 |
| 2 Feb 2025 | Chapters I and II apply: the prohibited AI practices of Article 5 and the AI literacy obligation of Article 4 | In application since 2 February 2025 | EUR-Lex, Regulation (EU) 2024/1689, Article 113(a) |
| 2 Aug 2025 | Obligations for general-purpose AI model providers (Chapter V), governance (Chapter VII), notified bodies (Chapter III Section 4), penalties except Article 101 (Chapter XII) and confidentiality (Article 78) apply | In application since 2 August 2025 | EUR-Lex, Regulation (EU) 2024/1689, Article 113(b) |
| 2 Aug 2026 | The Act's general application date: everything not given its own date, including the Article 50 transparency obligations; Article 101, under which the Commission can fine general-purpose AI model providers up to 3% of worldwide turnover or EUR 15 million; and Article 57's requirement that every member state have an operational AI regulatory sandbox | 2 August 2026 | EUR-Lex, Regulation (EU) 2024/1689, Articles 50, 57, 101 and 113 |
| 2 Aug 2027 | General-purpose AI models placed on the market before 2 August 2025 must be brought into compliance | Deadline, 2 August 2027 | EUR-Lex, Regulation (EU) 2024/1689, Article 111(3) |
| 2 Dec 2027 | Rules for high-risk AI systems in the Act's listed areas - including biometrics, critical infrastructure, education, employment, migration, asylum and border control - apply, as amended by the Digital Omnibus on AI | Applies from 2 December 2027; amendment in force since 27 July 2026 | EUR-Lex, Regulation (EU) 2026/1744, and the Commission's AI Act policy page, both linked below |
| 2 Aug 2028 | Rules for high-risk AI embedded in regulated products - such as lifts or toys, where AI is a safety component under EU product law - apply, as amended by the Digital Omnibus on AI | Applies from 2 August 2028; amendment in force since 27 July 2026 | EUR-Lex, Regulation (EU) 2026/1744, and the Commission's AI Act policy page, both linked below |
| 2 Aug 2030 | High-risk AI systems intended for use by public authorities that were already on the market before the rules applied must be brought into compliance | Deadline, 2 August 2030 | EUR-Lex, Regulation (EU) 2024/1689, Article 111(2) |
| 31 Dec 2030 | AI components of the EU's large-scale IT systems (the Act's Annex X) placed on the market before 2 August 2027 must be brought into compliance | Deadline, 31 December 2030 | EUR-Lex, Regulation (EU) 2024/1689, Article 111(1) |
What do most published EU AI Act timelines get wrong?
The most common error, as of the day we wrote this page, is presenting the Act’s original high-risk dates as current. Under the Act as adopted, high-risk rules were tied to the general application date of 2 August 2026 and to 2 August 2027.
Regulation (EU) 2026/1744 was proposed on 19 November 2025, politically agreed on 7 May 2026, adopted on 8 July 2026 and in force from 27 July 2026. Under it, the listed-area high-risk rules apply from 2 December 2027 and the product-embedded rules from 2 August 2028.
Timelines that predate the amendment, including carefully maintained ones, still showed the original dates when we compiled this page. The dates above are the amended ones, and the amending regulation is in the sources, so you can check this page the same way we checked the others.
What does this timeline not tell you?
It does not tell you what the law requires of your company. This is a dated record of when the Act’s parts apply, compiled from the regulation’s own text. It is not a compliance plan, a risk classification of any system, or legal advice, and Foxnut AI does not provide those. We state that boundary across the AI consulting pages at Foxnut AI.
Important note: the amendment also contains transitional arrangements beyond the headline dates recorded here. Where we could not verify a detail against primary text, we left it off this page and did not approximate it.
What the phases mean in practice for a company entering Europe is the companion page, the Act read from the operator’s side. Read that one next if you are trying to decide what to do and not just when.
How did we compile these dates?
As of July 2026, we have verified every row against a primary source, by reading the texts and not by citing citations. The dates from the Act itself come from Articles 111 and 113 of Regulation (EU) 2024/1689, read verbatim in the EUR-Lex full text. The amended high-risk dates come from Regulation (EU) 2026/1744 on EUR-Lex and the European Commission’s AI Act policy page as updated on 27 July 2026.
Where a secondary summary and the regulation text could disagree, the regulation text wins. All three sources are linked below, and each record names the specific article its date comes from, so you can re-verify any single row in minutes without trusting this page.
Who can you ask about the EU AI Act dates for your system?
You can ask us which of these dates touch the AI systems you run, as long as you take the legal answer to a lawyer. Foxnut AI is a practice of senior people with one tier, and the people who sign the proposal do the work. We designed, made and launched our own products before we advised anyone on theirs.
We build AI systems and hand them over running, so your own team can run them after we leave. The first step is a free expert audit, where a senior practitioner reviews your setup, with no RFP and no pitch deck. You can read about what Foxnut Consulting actually builds first.
If you are not sure which of these dates apply to your system, that is worth five minutes with us before it is worth a lawyer’s hour. Ask us.
Sources
- Regulation (EU) 2024/1689 (the AI Act), full text on EUR-Lex: Article 111 transitional provisions and Article 113 entry into force and application, read verbatimRetrieved
- Regulation (EU) 2026/1744 (Digital Omnibus on AI) on EUR-Lex: adopted 8 July 2026, published 24 July 2026, in force 27 July 2026, amending the high-risk application datesRetrieved
- European Commission: AI Act policy page, application timeline as amended (page updated 27 July 2026)Retrieved
Foxnut AI takes on work like this from Bengaluru and Paris. If you want to see how we do it before you talk to anyone, here is what an AI engagement costs before you ask.