EU Launches Review of Unpublished AI Models Under AI Act
The European Commission’s AI Office has initiated its first formal supervisory action under the EU AI Act, issuing requests for information regarding systemic risks to OpenAI, Anthropic, Google, and approximately 30 other AI providers.
What happened?
The European Commission and its AI Office have sent formal requests for information (RFI) to several leading developers of general-purpose AI (GPAI) models. Recipients include OpenAI, Anthropic, and Google, as well as over 30 other global AI providers. The regulatory action focuses on systemic risks and also encompasses AI models that have not yet been made public.
Key facts
| Datum för bekräftelse | 29 augusti 2026 |
|---|---|
| Tillsynsbefogenhet trädde i kraft | 2 augusti 2026 |
| Antal berörda AI-leverantörer | Över 30 leverantörer |
| Huvudfokus för RFI | Systemrisker och opublicerade GPAI-modeller |
Why it matters
This is the first confirmed supervisory action taken by the European Commission since it received formal enforcement powers over GPAI models under the AI Act on 2 August 2026. The fact that the RFI requests are also directed at unpublished models demonstrates the EU's intention to scrutinise systemic risks and autonomy before models are brought to market.
Who is affected?
The measure affects global AI developers, advanced research laboratories, and companies building products based on GPAI models for the European market. End users and European firms are indirectly affected by the tightening of transparency and safety assessment requirements for technology supplied to the EU.
Impact on the EU
The European Commission and its AI Office assumed direct supervision of GPAI models on 2 August 2026. The first formal RFI focuses specifically on systemic risks and applies to both Western and international actors whose models have reached or are expected to reach the EU market.
What else you should know
In the requests, the Commission is seeking detailed information regarding model architecture, training data, and safety testing. The report states that the investigations follow a series of incidents during the summer of 2026, during which advanced models were identified with potential autonomy and control risks.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka bolag berörs?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "EU Launches Review of Unpublished AI Models Under AI Act"