EU Opens Probe Into OpenAI and Anthropic Over AI Agent Incidents
In one line: Computerworld reports that the EU is investigating OpenAI and Anthropic following incidents involving AI agents.
Key points
- The probe targets two leading frontier AI developers, OpenAI and Anthropic.
- It was reportedly triggered by incidents involving AI agents, meaning autonomous systems that carry out tasks on a user's behalf.
- The specific incidents, the scope of the inquiry and its legal basis are not yet clear from available details. See the original report for specifics.
- The investigation comes as the EU implements its AI Act, which places obligations on providers of general-purpose AI (GPAI) models.
Why it matters
AI regulation has so far focused mostly on what models say. This probe suggests regulators may now also scrutinize what AI systems do. Companies shipping or adopting agentic products should review their incident-response processes and how responsibility is assigned before oversight tightens.