The Digital Omnibus proposes to postpone the high-risk obligations for standalone Annex III systems from 2 August 2026 to 2 December 2027, and to soften the wording of Article 4 AI literacy into a duty to institutionally promote and encourage literacy with proportionate measures. Article 50 transparency is not postponed and continues to apply from 2 August 2026. The European Parliament endorsed the text on 16 June 2026, but as of late June 2026 the package has not yet been formally adopted or published in the Official Journal. Until then, Regulation (EU) 2024/1689 in its current form remains the binding law.
The practical message is calm but clear. Plan against the original dates for as long as the Omnibus is not yet in force. Use the expected breathing room to get ahead of your obligations, not to pause your preparation. Below you will read exactly what the Omnibus proposes to change, what its legal status is, and what that means for your organisation's planning.
What does the Digital Omnibus propose to change?
The Digital Omnibus is a package with which the European Commission aims to simplify digital legislation and align it more closely with practice. For the AI Act it comes down to three core points: postponing part of the high-risk obligations, softening the duty around AI literacy, and deliberately leaving the transparency obligations untouched.
The postponement follows the two routes by which a system can be high-risk. The first route runs through Article 6(2) in combination with Annex III: standalone AI systems in sensitive domains such as recruitment, credit scoring, education, biometrics, critical infrastructure and law enforcement. For this category the date of application shifts from 2 August 2026 to 2 December 2027. The second route runs through Article 6(1) in combination with Annex I: AI embedded as a safety component in regulated products, such as medical devices. For those systems the date shifts to 2 August 2028. The fundamental rights impact assessment of Article 27, the FRIA, follows the high-risk timeline and moves along to 2 December 2027.
For Article 4 AI literacy, the Omnibus softens the wording. Where the current text speaks of ensuring a sufficient level of literacy, the proposal moves towards a mandate to promote and encourage literacy institutionally, with proportionate measures that suit the organisation. The duty itself does not disappear, but the hard penalty pressure is removed. For your approach this means: stay away from pure fear strategies and choose a line of empowerment, risk reduction and alignment with what your sector expects.
The third point is what the Omnibus deliberately does not do. The transparency obligations under Article 50, such as making clear that someone is talking to a chatbot and labelling deepfakes and synthetic content, are not postponed. They simply take effect on 2 August 2026.
What is the legal status as of late June 2026?
This is the part most often misunderstood. A political agreement and a vote in the European Parliament are important steps, but they do not yet make binding law.
State of play on 28 June 2026: The Commission published the Digital Omnibus on 19 November 2025. On 6 and 7 May 2026 the co-legislators reached a provisional political agreement. The European Parliament endorsed the text on 16 June 2026. But as of late June 2026 the amending regulation has not yet been formally adopted and has not yet been published in the Official Journal. Until that publication, the original AI Act and its timeline remain legally valid.
In concrete terms this means that the new dates mentioned in this article, such as 2 December 2027, are proposed dates that become binding only once the amendment enters into force. As long as that has not happened, 2 August 2026 is still the applicable date for the high-risk obligations under Annex III. Anyone who now stretches their planning based on a date that is not yet in law takes a risk that is easy to avoid.
The timeline at a glance
The table below sets the current, legally binding dates next to the dates the Digital Omnibus proposes, with the status alongside. Use the Current AI Act column for your planning for as long as the amendment is not yet published.
| Obligation | Current AI Act | Digital Omnibus proposal | Status |
|---|---|---|---|
| Prohibited practices (Art. 5) | 2 February 2025 | 2 February 2025 | Already applies |
| AI literacy (Art. 4) | 2 February 2025 | 2 February 2025 (softened wording) | Already applies |
| GPAI model obligations | 2 August 2025 | 2 August 2025 | Already applies |
| GPAI full enforcement | 2 August 2026 | 2 August 2026 | Not postponed |
| Transparency (Art. 50) | 2 August 2026 | 2 August 2026 | Not postponed |
| Art. 50 marking of existing generative AI | 2 August 2026 | 2 December 2026 | Limited grace period |
| High-risk Annex III (Art. 6(2)) | 2 August 2026 | 2 December 2027 | Proposed postponement |
| FRIA (Art. 27) | 2 August 2026 | 2 December 2027 | Follows high-risk |
| High-risk Annex I (Art. 6(1)) | 2 August 2027 | 2 August 2028 | Proposed postponement |
What applies regardless?
The postponement is selective. Three blocks do not move and therefore call for action now.
Prohibited practices (Article 5) have applied since 2 February 2025 and stand entirely apart from the high-risk timeline. The Omnibus in fact adds a new prohibition on non-consensual intimate imagery, with a transition period until 2 December 2026.
GPAI model obligations have applied since 2 August 2025, with the final GPAI Code of Practice of 10 July 2025 as a practical guide. From 2 August 2026 the Commission and the AI Office gain full enforcement powers, with a fine via the supervisory authority of up to 3 percent of worldwide annual turnover or 15 million euro. Existing GPAI models placed on the market before 2 August 2025 have until 2 August 2027 to comply.
Transparency obligations (Article 50) apply from 2 August 2026 and are not touched by the Omnibus. For the machine-readable marking of AI content there is a grace period until 2 December 2026 for systems already on the market. Anyone deploying chatbots, deepfakes or synthetic content should prepare for this now. For a step-by-step approach, see the transparency obligations checklist.
What does this mean for your planning?
The core is simple: do not let a proposed date stall your work. Until the Omnibus is in the Official Journal, plan against 2 August 2026 for the high-risk obligations. If the postponement arrives, use the extra months to work more thoroughly and carefully, not to begin.
Plan against the original date
Keep 2 August 2026 as your working date for your high-risk systems until the amendment is formally published. That way you never fall behind the applicable law.
Separate what applies regardless from what may be postponed
Map out which of your systems fall under Article 5, Article 50 or the GPAI obligations. Those blocks call for action now, regardless of the Omnibus.
Use the postponement to get ahead
If the high-risk date shifts to 2 December 2027, use that room for a thorough inventory, risk classification and a FRIA that holds up, not to delay the start.
Build evidence now
Supervision looks at what you can demonstrably show you have done. Record your AI inventory, role determination and AI literacy as you work, so the evidence builds itself.
The postponement is not a pause button. It is time you can invest in a file that a supervisory authority, a client or an internal audit can follow in a short space of time.
How do you put this into practice?
For execution it helps to separate the two tracks: governance and evidence on the organisation side, and people and literacy evidence on the side of your staff.
Embed AI runs an AI governance scan and a Readiness Sprint with which you order scope, AI inventory, risk classification and evidence around the right dates. Because the scan places the current and the proposed timeline side by side, you know which obligations apply now and which may shift, without pinning yourself to a date that is not yet in law.
LearnWize makes the people side demonstrable: per role which level of AI literacy is appropriate, with assessments, learning paths and training records that together form the evidence. Precisely now that the Omnibus softens the wording of Article 4 towards promoting and encouraging, it is wise to approach literacy as empowerment and to keep the evidence in order, because the duty itself remains.
For the legal background on the postponement, read the explainer on the postponement of the high-risk obligations to December 2027, and for the state of the political process the status of the Digital Omnibus.
Frequently asked questions about the Digital Omnibus and the AI Act
Short, citable answers for organisations aligning their AI Act planning with the Digital Omnibus.