Get the weekly explainer
EU and World AI Rules

EU AI Act news 2026: what changed and what comes next

What changed in the EU AI Act in 2026, in date order, with the next deadlines and links to the Official Journal and the Commission's own pages. Last checked 24 October 2026.

Published OCT 24, 2026Last checked OCT 24, 20264 min read
The short answer

The biggest EU AI Act news of 2026 is the Digital Omnibus on AI, published in the Official Journal on 24 July 2026. It moved most high-risk duties to 2 December 2027 and 2 August 2028 and added two new bans that start on 2 December 2026. Enforcement of the bans, transparency rules and general-purpose AI rules began on 2 August 2026. A separate Digital Omnibus covering the GDPR is still being negotiated.

01Where does the EU AI Act stand on 24 October 2026?

The AI Act is in force and has been amended once. The amending text, Regulation (EU) 2026/1744, is listed on EUR-Lex as in force. The Commission's timeline now shows the final roll-out running until 2 August 2028.

The next fixed date is 2 December 2026. This page is a dated summary, not breaking news and not legal advice. Ask a lawyer or your national authority before acting on a deadline.

02What happened this year, in date order?

  • 19 November 2025: the Commission proposed the AI part of the Digital Omnibus, according to the Parliament's legislative train.
  • 13 March 2026: the Council agreed its negotiating position.
  • 7 May 2026: Parliament and Council negotiators reached a provisional deal.
  • 16 June 2026: Parliament approved the text, with 423 votes for, 57 against and 174 abstentions.
  • 29 June 2026: the Council adopted it. The regulation was signed on 8 July 2026.
  • 24 July 2026: it was published in the Official Journal as Regulation (EU) 2026/1744.
  • 2 August 2026: enforcement started for the bans, transparency rules and general-purpose AI rules, as the Commission's FAQ explains.
  • 1 September 2026: a law firm's note on a Commission press briefing says the Commission confirmed its first information requests, sent to more than 30 AI companies on safety and security, and on copyright and transparency.

03What did the July 2026 amendment change?

The Commission's timeline and the amending regulation point to these main changes:

  • High-risk AI listed in Annex III now applies from 2 December 2027, and high-risk AI built into regulated products from 2 August 2028.
  • Two new bans apply from 2 December 2026, on AI that generates realistic intimate images of an identifiable person without consent and on AI that generates child sexual abuse material.
  • Certain providers already on the market before 2 August 2026 have until 2 December 2026 to meet the duty to mark AI-generated content.
  • The AI literacy duty in Article 4 now asks providers and deployers to take measures, and says they do not have to guarantee a set level of literacy for any person.
  • A new Article 4a gives providers and deployers a legal basis to process sensitive personal data to detect and correct bias, with safeguards.
  • Small mid-cap companies get some relief that small and medium companies already had, including lower fine ceilings.
  • The definition of a safety component is narrower, and the AI Office has wider exclusive supervision.

04Which guidance and codes have appeared?

The Commission's FAQ says a Code of Practice for the transparency rules was recently published. A law firm note reports that the Code was published on 10 June 2026, that the Commission and the AI Board found it adequate on 8 and 9 July, and that final guidelines on Article 50 followed on 20 July 2026 (Reed Smith). The code is voluntary.

Per the same FAQ, the Commission assessed the general-purpose AI Code of Practice as adequate, and signing up is optional. The guidelines on classifying high-risk AI were still a draft on the Commission's guidelines page, with consultation open until 23 July 2026.

Careful with this

The Commission's own pages on the transparency code and guidelines blocked our automated fetch, so those three dates come from a law firm's note. Check the Commission's pages before relying on them.

05What comes next?

Watch also for final guidelines on classifying high-risk AI, which are meant to help providers and deployers decide whether a system is in scope. To check for yourself each week, read the Commission's timeline and FAQ and the EUR-Lex page for the amending regulation. Compare their dates with this page's last checked date.

  • 2 December 2026: the two new bans and the transition for marking AI-generated content end.
  • 2 August 2027: each member state should have at least one AI regulatory sandbox.
  • 2 December 2027 and 2 August 2028: the high-risk rules apply.
  • The Digital Omnibus on data and the GDPR is a separate file. The Parliament's procedure page showed it as awaiting a committee decision, with a committee draft report published on 22 June 2026.

?What is not known

  • Whether and when the Commission will finalise its high-risk classification guidelines.
  • How the AI Office will publish enforcement outcomes. Sources we read mention information requests, not penalties.
  • What the separate GDPR omnibus will contain once it passes. Its text can still change.

QQuestions people ask

How do I keep up with EU AI Act news?

Use primary pages: the Commission's AI Act Service Desk timeline and FAQ, and EUR-Lex for the legal texts. Law-firm notes help, but treat them as secondary.

Has anyone been fined under the AI Act?

We have not found a published fine as of 24 October 2026. The AI Office can fine general-purpose model providers up to 3 percent of global annual turnover, according to the Commission's FAQ.

Sources and last checked

Last checked October 24, 2026. Rules and settings change, so open the source to see the current version.

  1. Regulation (EU) 2026/1744 EUR-Lex, Publications Office of the European Union The amending regulation; used for its date, number and publication.
  2. Timeline for the implementation of the EU AI Act European Commission, AI Act Service Desk Official dates after the amendment.
  3. AI Act frequently asked questions European Commission, AI Act Service Desk Used for enforcement start, codes of practice and the Omnibus changes.
  4. Article 4: AI literacy European Commission, AI Act Service Desk Consolidated text showing the amended literacy duty.
  5. Digital Omnibus on AI, legislative train European Parliament Procedure dates and the Parliament vote.
  6. Guidelines on the classification of high-risk AI systems European Commission, AI Act Service Desk Shows the draft status and consultation date.
  7. Digital Omnibus procedure 2025/0360(COD) European Parliament, Legislative Observatory Stage of the separate data and GDPR omnibus.
  8. Transparency obligations for AI-generated content Reed Smith (law firm) Secondary source for the code and guideline dates.
  9. EU AI Act enforcement has started Allegiance Law (law firm) Secondary source for the first information requests.

General information, not legal, medical, financial or investment advice. We are not affiliated with any company, agency or standards body named here. See our editorial policy and corrections, or tell us about a mistake at [email protected].