You are not logged in so some information on this page has been withheld. To see more, please log in or sign up.

2026 › August › 05

AI Safety Institute

9 days ago

AI used new levels of 'autonomy and deception' to trick people in safety test ⊕

found 2026 › Aug › 05 (Wed) @ 00:15 UTC

a story from BBC ✅ › BBC News

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

page 1 of 1