Three tracks. One governance posture.
The AI Omnibus, Regulation (EU) 2026/1744, in force 27 July 2026, leaves three tracks running at different speeds: Article 50 transparency, Annex III high-risk and Annex I high-risk. The two high-risk tracks demand the same underlying evidence, listed below; the dates are in the timeline.
No compliance claim under these frameworks. iTmethods makes no compliance, certification or accreditation claim under the EU AI Act or any other framework named on this page. Beyond SOC 2 Type II for Reign Ops, no other certification or accreditation is claimed.
More on what we do not claim
We are not certified against, accredited for, or approved under any of them. Alignment is not compliance. The determination belongs to your own risk function, and nothing here transfers it to us. No authority responsible for it endorses, approves or recommends Reign or iTmethods.
Every high-risk obligation lands on the same evidence layer.
The obligations differ by track. What they ask you to produce differs a good deal less.
Classification
Which of your systems are high risk, on what basis, and who decided. The answer changes as use cases move, so it is a record rather than a one-time assessment.
Could you defend your classification?
Documentation and oversight
Technical documentation, human oversight and record keeping. Three obligations that all resolve to the same question: what happened, and can you show it.
Show it from where?
Monitoring after deployment
The obligation does not end at go-live. Post-market monitoring is continuous, and evidence assembled afterwards is the weakest kind.
What is watching yours today?
Reign Gateway is the relevant product for policy at the model boundary and for logging each crossing. The obligations stay with your institution. Reign Gateway carries the operating record they are assessed against.
The deadline moved. The work did not.
The AI Omnibus, Regulation (EU) 2026/1744, in force 27 July 2026, extends the date. It does not reduce what has to exist by that date.
| Obligation | Date | Status | Moved by the Omnibus? |
|---|---|---|---|
| Prohibitions 1 to 8 and AI literacy | February 2025 | Applies now | No |
| General-purpose AI (GPAI) obligations | August 2025 | Applies now | No |
| Article 50 transparency obligations (the bulk) | 2 August 2026 | Applies now | No |
| Synthetic-content marking, for systems on the market before 2 August 2026 | 2 December 2026 | Still ahead, next | No |
| Prohibition 9, introduced by the AI Omnibus | December 2026 | Still ahead, next | Added by the Omnibus |
| Annex III high-risk (employment, credit scoring, education, essential services) | 2 December 2027 | Still ahead | Yes, previously 2 August 2026 |
| Annex I high-risk (AI embedded in regulated products, including medical devices) | 2 August 2028 | Still ahead | Yes, previously 2 August 2027 |
Prohibition 9 is on the Commission’s own AI Act page, which says: “Prohibition 9 comes into effect in December 2026 and was introduced as a part of the AI Omnibus.” The Commission names the month only, so this table does the same. The Status column is as of 27 October 2026, and it is next checked on 3 December 2026, the day after the 2 December 2026 date.
What did not move. The bulk of the Article 50 transparency obligations have applied since 2 August 2026. Systems already on the market before that date have until 2 December 2026 to meet the synthetic-content marking requirement. Both dates come before the high-risk deadlines in 2027 and 2028. If any part of this timeline is doing work for you today, it is this part.
Institutions building the governance posture now will have more of the evidence already in place. Whether that amounts to readiness is their risk function’s determination. Institutions that wait face the same workload on a shorter clock, and the parts that take longest are the parts that cannot be bought at the end.
Start on the evidence.
We will look at which obligations your current estate could already answer, and which have nothing behind them yet.