Urgent.News

What's breaking now, across thousands of outlets.

AI

AI companies are debating whether to take cyber tests offline after their models hacked each other

Models from OpenAI, Anthropic, and Meta each compromised outside systems during security evaluations, prompting calls for new industry standards

AI companies are debating whether to take cyber tests offline after their models hacked each other

We haven't written up this one. Quartz has the full story — the link below goes straight to it.

Read the original at qz.com →

More in AI

More from Tuesday 25 August →