Digital Omnibus status at a glance
The Digital Omnibus on AI is a short amending regulation. It does not replace the EU AI Act; it edits it. Its formal title is Regulation (EU) 2026/1744 amending Regulations (EU) 2024/1689, (EU) 2018/1139 and (EU) 2023/1230 as regards the simplification of the implementation of harmonised rules on artificial intelligence. The text is on EUR-Lex.
The European Commission proposed it on November 19, 2025 as part of a wider Digital Package on Simplification. Parliament and Council reached a political agreement on May 7, 2026, and the regulation entered into force on July 27, 2026, according to the Commission's AI Act page. The act was signed on July 8, 2026 and published in the Official Journal on July 24, 2026 (White & Case). It took effect on the third day after publication instead of the usual twentieth so that the new dates would be law before August 2, 2026, when the high-risk rules were originally due to apply.
The regulation has 47 recitals and four articles, and it adds a new Annex XIV to the AI Act. Article 1 amends the AI Act, Article 2 amends the civil aviation regulation and Article 3 amends the Machinery Regulation, according to NicFab's article-by-article analysis of the published text.
What the Digital Omnibus changed in the EU AI Act
The table sets out each date the Omnibus moved or added, with the provision of the amended AI Act that now fixes it. Dates are as of November 10, 2026.
| Provision | Original date | New date | Who it affects | Reference in the amended AI Act |
|---|---|---|---|---|
| High-risk obligations, Annex III systems (Chapter III, Sections 1 to 3) | August 2, 2026 | December 2, 2027 | Providers and deployers of AI used in biometrics, critical infrastructure, education, employment, essential services, law enforcement, migration and justice | Article 113, third paragraph, point (c)(i) |
| High-risk obligations, Annex I systems | August 2, 2027 | August 2, 2028 | Makers of regulated products with AI safety components, such as medical devices, toys and lifts | Article 113, third paragraph, point (c)(ii) |
| Machine-readable marking of generated content, systems already on the market | August 2, 2026 | December 2, 2026 | Providers of generative AI systems placed on the market before August 2, 2026 | Article 111(4); Article 50(2) |
| New prohibition: non-consensual intimate content and child sexual abuse material | None (new) | December 2, 2026 | Providers and deployers of systems that generate such material | Article 5(1), points (ba) and (bb); Article 113, third paragraph, point (a) |
| National AI regulatory sandboxes | August 2, 2026 | August 2, 2027 | Member states | Article 57(1) |
| Application for AI Act designation by bodies already notified under product law | None (new) | January 28, 2028 | Notified bodies under Annex I, Section A legislation | Article 43(3) |
The original and new dates in the first five rows match Cooley's August 2026 summary; the article references and the notified body deadline are from NicFab's analysis.
Changes beyond the dates
- AI literacy softened. Article 4 now requires providers and deployers to take measures to support AI literacy among staff, instead of ensuring a sufficient level of it. The new text applies from July 27, 2026 with no deferral.
- Safety component narrowed. AI functions that only assist users or optimize performance are not automatically safety components if their failure creates no health or safety risk, which keeps some products out of the high-risk category.
- Machinery moved. The Machinery Regulation moves from Section A to Section B of Annex I, so AI in machinery is governed through sectoral rules and not the full high-risk regime.
- SME relief extended. Simplified documentation and other relief for small and medium-sized enterprises now also cover small mid-cap companies.
- AI Office powers expanded. The Commission's AI Office gains exclusive competence over AI systems built on a general-purpose model by the same provider or corporate group, and over AI that is part of a very large online platform or search engine under the Digital Services Act.
- Registration kept. The Commission had proposed dropping EU database registration for systems a provider self-assesses as not high-risk under Article 6(3). The co-legislators kept the duty, with reduced information.
These points are drawn from the White & Case and Cooley alerts on the final text.
Is the EU AI Act delayed?
Partly. The Omnibus delayed one block of the Act: Sections 1, 2 and 3 of Chapter III, which hold the classification rules, the requirements for high-risk systems and the obligations of their providers and deployers. Everything else kept its date. The Act's general date of application, August 2, 2026, did not change.
That is why "EU AI Act delayed" is accurate for a company building hiring software and wrong for a company running a chatbot. The Commission's own summary says the Act became applicable on August 2, 2026 with exceptions, and lists the high-risk extension as one of them (European Commission).
Entry into force and application are also different things. On July 27, 2026 the amendments became part of the AI Act's text, but each amended provision becomes enforceable on the date that Article 113 sets for its chapter. The softened Article 4 sits in a chapter that has applied since February 2025, so it took effect at once. The amended high-risk articles follow the 2027 and 2028 dates.
When do the EU AI Act high-risk rules apply now?
December 2, 2027: Annex III systems
Annex III lists the stand-alone uses the Act treats as high-risk: biometrics, critical infrastructure, education, employment, access to essential private and public services, law enforcement, migration and border control, and the administration of justice. Providers of these systems were due to comply by August 2, 2026. They now have until December 2, 2027, a deferral of 16 months.
August 2, 2028: Annex I systems
Annex I covers AI that is a safety component of a product already regulated by EU product safety law, or is itself such a product. The original date was August 2, 2027. The new date is August 2, 2028, a deferral of 12 months.
Why fixed dates
The stated reason for the delay is that the harmonised technical standards companies need to show conformity were not ready. The Commission's November 2025 proposal would have tied the new dates to a later decision that standards and guidance were available. The final text dropped that trigger and set fixed dates. The Commission's AI Act page lists a first harmonised standard supporting the Act as published in the Official Journal in October 2026 (European Commission).
Systems already on the market
Two transition rules sit alongside the new dates. Under Article 111(2), a high-risk system already on the market before its compliance date is generally caught only if its design later changes significantly, and Cooley notes this now operates at the level of type and model. High-risk systems intended for use by public authorities must comply by August 2, 2030, a date the original Act already set and the Omnibus left unchanged.
What did not move
Four parts of the Act are on their original schedule, and three of them already apply.
- Prohibited practices. The eight Article 5 bans have applied since February 2, 2025.
- General-purpose AI models. The documentation, copyright policy and systemic-risk duties in Chapter V have applied since August 2, 2025.
- Transparency. Article 50, which covers chatbot disclosure, machine-readable marking of generated content and deepfake labels, has applied since August 2, 2026. The only change is the four-month grace period, to December 2, 2026, for marking by systems that were already on the market. Our Article 50 guide covers these duties.
- Governance and enforcement. The AI Office's enforcement powers over general-purpose models and the national authority framework have applied since August 2, 2026.
The Omnibus is the AI strand of a larger package. White & Case noted in July 2026 that the package's proposed changes to the GDPR, the ePrivacy Directive, NIS2 and the Data Act were still being negotiated and were not law (White & Case). This page covers only the AI regulation.
The new ban is a tightening, not a delay
The Omnibus adds a ninth prohibited practice. From December 2, 2026 it is unlawful to place on the EU market an AI system whose intended purpose is to generate realistic intimate or sexually explicit depictions of an identifiable person without consent, or child sexual abuse material. The ban also reaches systems where that output is a reasonably foreseeable result of the design and the provider lacks adequate safeguards. Deployers are covered where they use a system for that purpose. There is no exemption for systems already on the market (Cooley).
What this means for providers and deployers
The practical effect depends on which part of the Act a company falls under.
| If you | Your next date | What applies |
|---|---|---|
| Provide a generative AI system launched before August 2, 2026 | December 2, 2026 | Machine-readable marking under Article 50(2) |
| Provide or deploy any image, video or audio generator | December 2, 2026 | The new ban on non-consensual intimate content and child sexual abuse material |
| Provide a general-purpose AI model placed on the market before August 2, 2025 | August 2, 2027 | Full compliance with the general-purpose model rules (Article 111) |
| Provide or deploy an Annex III high-risk system | December 2, 2027 | Chapter III requirements, conformity assessment, registration |
| Make a regulated product with an AI safety component | August 2, 2028 | Chapter III requirements through the product's conformity route |
Law firm guidance on the final text is consistent on one point: the extra time is for building compliance, not pausing it. Cooley describes the Omnibus as targeted relief and not a general reset, and White & Case advises companies to update their roadmaps to the new dates while prioritizing the duties that applied from August 2, 2026. For US companies the reach of the Act is unchanged, since it applies to providers that place systems on the EU market wherever they are established. Other national regimes are compared on our international AI laws page.
The Digital Omnibus, in force since July 27, 2026, gives providers and deployers of high-risk AI more time: December 2, 2027 for Annex III systems and August 2, 2028 for Annex I products. It did not delay the rest of the EU AI Act. The bans, general-purpose model rules and Article 50 transparency rules apply now, and two December 2, 2026 dates are close: the end of the marking grace period and the new nudifier and CSAM ban.
What changed
- 2026-11-15: Page published, with dates verified as of November 10, 2026.
- 2026-08-02: AI Act became generally applicable; Article 50 and enforcement powers began to apply on the original schedule.
- 2026-07-27: Regulation (EU) 2026/1744 entered into force.
- 2026-07-24: Regulation (EU) 2026/1744 published in the Official Journal.
- 2026-07-08: Regulation signed in Strasbourg.
- 2026-05-07: Political agreement reached between Parliament and Council.
- 2025-11-19: Commission proposed the Digital Omnibus on AI.
Frequently asked questions
What is the Digital Omnibus?
The Digital Omnibus on AI is Regulation (EU) 2026/1744, an amending law that changes parts of the EU AI Act. It delays the high-risk obligations, adds a ban on nudifier apps and AI-generated child sexual abuse material, softens the AI literacy duty and widens the AI Office's powers. It is one strand of the Commission's wider digital simplification package.
Is the EU AI Act delayed?
Only the high-risk obligations are delayed. Annex III systems moved from August 2, 2026 to December 2, 2027, and Annex I product systems from August 2, 2027 to August 2, 2028. The bans, general-purpose AI model rules, Article 50 transparency rules and enforcement framework kept their original dates and apply now.
When do the EU AI Act high-risk rules apply now?
From December 2, 2027 for stand-alone high-risk uses listed in Annex III, such as hiring, credit and education systems, and from August 2, 2028 for AI that is a safety component of products regulated under Annex I, such as medical devices and toys. These are fixed dates, not conditional on standards being ready.
When did the Digital Omnibus take effect?
It was published in the Official Journal on July 24, 2026 and entered into force on July 27, 2026, the third day after publication. Individual amendments apply on the dates set in Article 113 of the AI Act, so the new prohibition, for example, applies from December 2, 2026.
Sources
- EUR-Lex: Regulation (EU) 2026/1744 (Digital Omnibus on AI)
- EUR-Lex: Regulation (EU) 2024/1689 (AI Act), Official Journal
- European Commission: AI Act policy page, application timeline and simplification
- White & Case: EU AI Omnibus enters into force, amending the AI Act (July 2026)
- Cooley: Digital AI Omnibus delays key deadlines, introduces new rules (Aug. 3, 2026)
- NicFab: Regulation (EU) 2026/1744 is published in the Official Journal (July 24, 2026)
This page is journalism and commentary, not legal advice. AI law changes quickly; check the primary sources above and consult qualified counsel for your situation.