Has the EU AI Act high-risk deadline been postponed?

Yes, for the high-risk obligations. Under the Digital Omnibus on AI — now Regulation (EU) 2026/1744, in force since 27 July 2026 — the obligations for stand-alone high-risk systems (Annex III) move from 2 August 2026 to 2 December 2027, and those for AI embedded in regulated products (Annex I) move from 2 August 2027 to 2 August 2028. The change is now in force.

Does anything in the EU AI Act still apply on 2 August 2026?

Yes. The Digital Omnibus does not defer the Article 50 transparency obligations — chatbot disclosure, and the marking and labelling of AI-generated audio, image, video, and text. It also does not touch the Article 5 prohibitions or the general-purpose AI model rules that already apply. Only the high-risk obligations were moved.

Why were the high-risk AI Act deadlines delayed?

The stated reason is sequencing: to ensure the technical standards and support tools that companies need to comply are actually in place before the obligations start to apply. The new dates are fixed calendar dates, not conditional triggers.

Should we stop preparing for high-risk AI Act compliance?

No. The obligations were postponed, not removed, and building a risk-management process, data governance, documentation, and human oversight for a high-risk system takes many months. The extra time is runway, not a reprieve.

Has the EU AI Act high-risk deadline been delayed to 2027?

If your company has been planning around 2 August 2026 as the day the EU AI Act's high-risk obligations bite, that date has moved. A recent amendment — the Digital Omnibus on AI — postpones the high-risk timeline by more than a year. But it does not postpone everything, and the parts it leaves in place still land on 2 August 2026. Here is what actually changed.

This is an information service to help you plan, not legal advice. For the obligations tied to your own AI use, see which apply to your company.

What the Digital Omnibus did

The Digital Omnibus on AI is an amendment to the AI Act aimed at simplification and sequencing. It was adopted as Regulation (EU) 2026/1744, published in the Official Journal on 24 July 2026, and entered into force on 27 July 2026.

Its headline effect is to delay the application of the high-risk obligations. It does not repeal them, and it does not touch the prohibitions or the transparency rules.

The new high-risk dates

High-risk category	Old application date	New application date
Stand-alone high-risk systems (Annex III) — employment, credit and essential services, education, biometrics, critical infrastructure, migration, justice	2 August 2026	2 December 2027
High-risk AI embedded in regulated products (Annex I) — medical devices, machinery, toys, and similar	2 August 2027	2 August 2028

The stated rationale is sequencing: the change is meant to ensure the harmonised technical standards and other support tools that companies rely on to comply are in place before the obligations apply. The new dates are fixed calendar dates — not triggers that depend on the standards actually being finished.

For the full phased schedule, see the EU AI Act timeline — note that the high-risk milestones on that calendar are the ones this amendment moves.

What still applies on 2 August 2026

This is the part that gets lost in the "deadline delayed" headlines. The Digital Omnibus moved the high-risk obligations. It did not defer:

Article 50 transparency obligations. You must still tell people when they are interacting with an AI system (chatbots), and providers must mark AI-generated or manipulated audio, image, video, and text so it is detectable as such; deepfakes and certain AI-generated text on matters of public interest must be labelled. To help, the Commission published a voluntary Code of Practice on marking and labelling AI-generated content on 10 June 2026.
The Article 5 prohibitions, which have applied since 2 February 2025 — and which the Omnibus actually extends, adding a new prohibition on AI systems that generate non-consensual intimate imagery of real people ("nudifier" tools).
The general-purpose AI (GPAI) model obligations, which have applied since 2 August 2025.

So if your only AI exposure is a customer-facing chatbot or AI-generated marketing content, your 2 August 2026 date has not moved. If you build or deploy a high-risk system, it has.

Is your system even high-risk?

The delay only matters to you if you are actually in the high-risk tier. That turns on Article 6 and Annex III — a specific test with an exemption most companies miss. Before you re-plan around 2 December 2027, confirm which side of the line you are on: is your AI system high-risk?

What this means for your planning

Don't down-tools. The obligations were postponed, not deleted. Risk management, data governance, technical documentation, logging, and human-oversight processes are built over quarters, not weeks — the runway to 2 December 2027 is there to be used.
Separate your two clocks. Transparency and prohibition duties on the 2 August 2026 timeline are independent of the high-risk clock now running to December 2027. Track them separately so a "we have more time" message on one doesn't quietly delay the other.
Work from the published text. The Digital Omnibus is now published as Regulation (EU) 2026/1744 and entered into force on 27 July 2026, so the new dates apply. As always, confirm the specifics against the official source before you commit them to a compliance plan.

The cleanest way to turn all of this into your own dated list: answer three questions about your AI use and we will show you the verified obligations that apply, sorted by the deadline you need to hit first. Because dates like these move, join the waitlist to be told the moment something that affects you changes.

The official text is Regulation (EU) 2024/1689; the amending act is Regulation (EU) 2026/1744 (the Digital Omnibus on AI). This article is an information service to help you orient — it is not legal advice, and you should confirm the current dates against the official source before acting.