Ok, Europe vs AI, part two.

Ok, Europe vs AI, part two.

An AI agent escapes its test environment and hacks another company. Sounds like a Netflix plot, but this summer it really happened, an OpenAI agent and Hugging Face. And an Anthropic model got unauthorised access to three outside organisations.

Now the EU AI Office sent both companies requests for information. Lawyers already say the AI Act does not cover testing at all, there is an R&D exemption. The Commission says if it escapes the lab and hits the EU market, it counts. They compare it to a virus escaping a lab..

Euronews had the whole story yesterday https://www.euronews.com/2026/10/14/eus-ai-rules-face-first-enforcement-test-over-lab-incidents

Last time I was complaining about the EU. This time I'm kind of on their side? Somebody has to ask these questions, no?