EU AI Act · Article 50 in force · marking deadline 2 December 2026
EU AI Act · Status and timeline

Is the EU AI Act in force?

Adopted 2024 · applies in stagesArticle 50 applies since 2 Aug 2026Last reviewed August 2026

Yes. The EU AI Act, Regulation (EU) 2024/1689, was published in the Official Journal on 12 July 2024 and entered into force on 1 August 2024. It applies in stages rather than all at once: prohibited practices since 2 February 2025, general-purpose AI rules since 2 August 2025, and, the stage most websites meet first, the Article 50 transparency obligations, which apply since 2 August 2026, with national enforcement powers live from the same day. The dates still ahead mostly concern high-risk systems, not everyday websites.

"In force" and "applies" are two different things, and most confusion about this law lives in the gap. A regulation enters into force once (here, 1 August 2024), which starts the clock; individual obligations then start applying on later, staggered dates. So "is the AI Act in force?" has been a yes since 2024, but the question people usually mean, "do its rules bind me yet?", depends on which rules. For the transparency duties that show up on a website (chatbot disclosure, AI-content labels), the answer became yes on 2 August 2026.

The full timeline, date by date

As amended by Regulation (EU) 2026/1744 (the Digital Omnibus, in force 27 July 2026), the staged application looks like this:

DateWhat starts applyingStatusas of August 2026
12 July 2024The Act is published in the Official Journal as Regulation (EU) 2024/1689.Done
1 August 2024Entry into force. No obligations apply yet; the implementation clock starts.Done
2 February 2025Prohibited AI practices (Article 5) and AI literacy duties (Article 4).Applying
2 August 2025General-purpose AI model obligations, governance bodies, the notified-body framework, and the penalties chapter. Deadline for member states to designate national authorities and set penalty rules.Applying
2 August 2026General application. Article 50 transparency duties: chatbot disclosure (50(1)), machine-readable marking of AI content (50(2)), emotion-recognition notice (50(3)), deepfake and AI-text labels (50(4)). National authorities can fine Article 50 breaches from this date. Regulatory sandboxes must be operational.Applying
2 December 2026The one Article 50 carve-out ends: machine-readable marking (50(2)) becomes due for generative systems that were already on the market before 2 August 2026. What this date does and does not cover.Ahead
2 August 2027General-purpose AI models placed on the market before 2 August 2025 must be fully compliant.Ahead
2 December 2027Stand-alone high-risk AI systems (Annex III: employment, education, credit and similar uses). Moved to this date by Regulation (EU) 2026/1744.Ahead
2 August 2028High-risk AI embedded in regulated products (Annex I: machinery, medical devices and similar). Moved to this date by Regulation (EU) 2026/1744.Ahead

Application dates per Article 113 of the Act as amended; high-risk dates reflect the July 2026 Digital Omnibus amendment.

What "applies" means for a website owner today

Since 2 August 2026, the duties a first-time visitor could observe on a live site are binding law:

Geography does not exempt a business: the Act covers providers and deployers outside the EU whenever the system's output is used in the Union, so US and UK companies with EU visitors are in scope. There is no general small-business exemption, only a lower fine cap.

What regulators can already do

Since 2 August 2026, national market-surveillance authorities can penalise Article 50 breaches with fines of up to €15 million or 3% of worldwide annual turnover, whichever is higher; for SMEs and start-ups the cap is whichever of those is lower (Article 99(4)(g) and 99(6)). The Commission's Article 50 guidelines of 20 July 2026 are the reference document those authorities are expected to assess against.

Two honest caveats belong next to that. First, enforcement machinery is uneven: member states were due to designate their authorities by August 2025, and as of March 2026 only 8 of 27 had notified their single points of contact, with several still designating. Second, as of August 2026 no Article 50 enforcement action has been publicly recorded anywhere in the EU. Neither caveat changes what the law requires, and the fines are ceilings a late-starting authority can still apply to conduct after 2 August 2026. What an authority (or a complaining competitor) checks is simple: they open the site, start the chat, look at the published content. Documented, dated evidence of what your site showed, and when, is what answers that kind of question, which is why we keep sealed records of every scan.

What the 2026 Digital Omnibus changed, and what it didn't

Regulation (EU) 2026/1744, published 24 July 2026 and in force since 27 July 2026, is the source of most "AI Act delayed" headlines. What it moved: the high-risk regime, to 2 December 2027 (Annex III) and 2 August 2028 (Annex I). What it did not move: Article 50, which applies from 2 August 2026 as originally scheduled. It also confirmed the single narrow transitional: machine-readable marking under 50(2), for generative systems already on the market before 2 August 2026, until 2 December 2026. Chatbot disclosure and content labels have no grace period.

This page describes the staged application as understood at the August 2026 review, including the amendments made by Regulation (EU) 2026/1744. Dates for parts of the Act not yet applying could move again; the statute text linked below is the authority.

Common questions

Is the EU AI Act in force?

Yes. The EU AI Act (Regulation (EU) 2024/1689) entered into force on 1 August 2024 and has been applying in stages since: prohibited practices and AI literacy duties since 2 February 2025, general-purpose AI rules since 2 August 2025, and the Article 50 transparency obligations, together with most remaining duties and the national enforcement powers behind them, since 2 August 2026.

When does the EU AI Act go into effect?

It already has. The Act entered into force on 1 August 2024, and its transparency obligations have applied since 2 August 2026. The dates still ahead concern other parts of the Act: 2 December 2026 (machine-readable marking catches up with generative systems that were on the market before August 2026), 2 August 2027 (legacy general-purpose AI models), 2 December 2027 (Annex III high-risk systems), and 2 August 2028 (high-risk AI in regulated products).

Has the EU AI Act been passed?

Yes, in 2024. It was published in the Official Journal on 12 July 2024 as Regulation (EU) 2024/1689 and entered into force on 1 August 2024. It has since been amended once, by Regulation (EU) 2026/1744 (the Digital Omnibus, July 2026), which moved the high-risk deadlines to 2027-2028 but left the transparency obligations applying from 2 August 2026.

What is the EU AI Act's effective date?

There is no single one, which is why answers to this question disagree. Entry into force was 1 August 2024. The date most businesses actually mean, when their own visible duties started, is 2 August 2026, when the Article 50 transparency rules and the general application of the Act began. High-risk obligations follow on 2 December 2027 and 2 August 2028.

Does the EU AI Act apply to the UK?

It is not UK law, but it can still reach a UK business. The Act expressly covers providers and deployers established outside the EU whenever their AI system's output is used in the Union, so a UK company with EU visitors, customers, or content published into the EU is in scope for the transparency duties in the same way a US company is.

Sources

This page summarises the law's status for orientation. It is not legal advice; where your situation is non-obvious, involve qualified counsel.

The law applies. Where does your site stand? DisclosureProof opens your site the way a first-time visitor would and records what an Article 50 check observes: chat-widget disclosure at first interaction, media marking, and labels, sealed into a dated evidence record. Run the free scan →