EU AI Act: current status and deadlines.
The EU AI Act applies in stages, and the timeline changed in 2026. See at a glance what applies today, what the Digital Omnibus postponed, who is in charge in Germany and which deadlines come next.
As of 28 October 2026 · checked monthly · not legal advice
Key points
- The AI Act has been in force since 1 August 2024 and applies in stages. Bans and AI literacy have applied since 2 February 2025, rules for general-purpose AI models since 2 August 2025.
- Since 2 August 2026, the Article 50 transparency duties have applied, for example to chatbots, regardless of risk class.
- The Digital Omnibus (Regulation (EU) 2026/1744, in force since 27 July 2026) moved high-risk duties to fixed later dates: 2 December 2027 and 2 August 2028.
- In Germany, the Bundesnetzagentur has been the market-surveillance authority, contact point and complaints office for the AI Act since 29 July 2026.
What already applies under the AI Act.
The AI Act entered into force on 1 August 2024, but its duties apply in stages. As of 28 October 2026, four stages already apply:
Prohibited practices (Art. 5)
Since 2 February 2025. Certain AI uses are banned in the EU.
Since 2 February 2025. Companies support their staff’s AI literacy.
AI literacy duty →General-purpose AI models
Since 2 August 2025 duties apply to providers of such models; since 2 August 2026 the Commission has enforcement powers.
Since 2 August 2026, for example for chatbots and AI-generated content. Not limited to high-risk; open-source systems are not exempt.
EU AI Act Article 50 →All EU AI Act deadlines at a glance.
| Date | What applies | Status 28 Oct 2026 |
|---|---|---|
| 1 August 2024 | AI Act enters into force | done |
| 2 February 2025 | Prohibited practices (Art. 5) and AI literacy (Art. 4) | applies |
| 2 August 2025 | Duties for general-purpose AI models, governance | applies |
| 2 August 2026 | Transparency duties (Art. 50) and Commission enforcement powers for general-purpose AI models | applies |
| 2 December 2026 | Two new prohibitions under Art. 5; end of the transition period for machine-readable marking (Art. 50(2)) for systems placed on the market before 2 August 2026 | upcoming |
| 2 December 2027 | High-risk systems under Annex III (stand-alone) | upcoming |
| 2 August 2028 | High-risk AI in regulated products under Annex I | upcoming |
Source: Regulation (EU) 2024/1689 as amended by Regulation (EU) 2026/1744 (Digital Omnibus). We check this page monthly.
What changed in the AI Act in 2026.
2026 was a year of change for the AI Act. Four changes affect most companies:
Digital Omnibus is law
Regulation (EU) 2026/1744, published 24 July 2026, in force since 27 July 2026. It fixes the high-risk dates and adds two prohibitions.
High-risk later
Stand-alone high-risk systems under Annex III from 2 December 2027, high-risk AI in regulated products from 2 August 2028: fixed dates instead of a link to standards.
Article 50 guidelines
The Commission adopted the final guidelines on the transparency duties on 20 July 2026. They are non-binding but are the supervisors’ working basis.
Supervision in Germany
The KI-MIG has been in force since 29 July 2026. The Bundesnetzagentur coordinates and runs an AI service desk and regulatory sandboxes.
Important for Germany: according to the federal digital ministry, the KI-MIG builds on the established responsibilities of the existing market-surveillance authorities, so companies keep their familiar contacts, including for AI Act questions. The Bundesnetzagentur coordinates, but does not replace every existing supervisor.
BMDS press release (29 July 2026, German) →EU AI Act overview →
What companies should do now.
Start with an overview: which AI systems does your company use, for what, and what needs checking for each? A simple AI inventory is enough to begin with.
| AI system | Purpose | To check |
|---|---|---|
| Website chatbot | customer questions | transparency (Art. 50) |
| Knowledge assistant | internal questions | risk class |
| Applicant tool | pre-selection | Annex III point 4 |
| Text generator | marketing | labelling (Art. 50) |
Map your AI use
Which AI systems does who use, for what and with which data?
Check the risk class
Is a use high-risk, for example in HR or lending?
Support AI literacy
Set training, policy and a contact person (Art. 4) and document the evidence.
Ensure transparency
Make chatbots and AI content recognisable under Art. 50.
Prepare for high-risk
Build documentation, oversight and processes by December 2027.
Where to follow changes.
Official Journal of the EU
The binding text of Regulation (EU) 2024/1689 and its amendments is on EUR-Lex.
Bundesnetzagentur
AI service desk with information and contacts for companies in Germany.
This page
We check the status monthly and note the date of the last check at the top.
This is general information, not legal advice. Have your specific use assessed by your legal adviser.
Questions about the status of the AI Act.
Is the AI Act already in force?
Yes. The AI Act has been in force since 1 August 2024, but its duties apply in stages. Bans and AI literacy have applied since 2 February 2025, rules for general-purpose AI models since 2 August 2025 and the Article 50 transparency duties since 2 August 2026.
When do the high-risk duties apply?
After the Digital Omnibus, from 2 December 2027 for stand-alone high-risk systems under Annex III, for example in HR or education, and from 2 August 2028 for high-risk AI in regulated products under Annex I. The dates are fixed and no longer linked to standards.
What did the Digital Omnibus change?
Regulation (EU) 2026/1744 has been in force since 27 July 2026. It moved high-risk duties to fixed later dates, added two new prohibitions from 2 December 2026 and, for systems placed on the market before 2 August 2026, created a transition period for machine-readable marking until 2 December 2026.
Who is responsible for the AI Act in Germany?
Since 29 July 2026, under the KI-MIG, the Bundesnetzagentur is the market-surveillance authority, contact point and complaints office. It coordinates and offers an AI service desk. Companies also keep their existing contacts at the established market-surveillance authorities.
What are the penalties for breaching the transparency duties?
For breaches of the Article 50 transparency duties, the AI Act provides for fines of up to €15 million or 3 per cent of worldwide annual turnover. Market-surveillance authorities have been able to enforce this since 2 August 2026. This is not legal advice.
Questions about your duties under the AI Act?
Stefan walks you through which deadlines matter for your AI use and what to document now.