Tech
Anthropic Tightens AI Controls
Following a series of incidents, Anthropic is cutting off internet access for all internal evaluations to prevent unintended model actions. Recent cases included submitting false tips about unsolved murders.
Coverage timeline — 1 article
The Verge
After a recent spate of high-profile incidents in which AI agents escaped containment, Anthropic is cutting off internet access for all internal evaluations. In a report Friday, the company detailed "unintended model actions," including submitting a false tip regarding an unsolved murder, that led t
2026-10-10 14:41 UTC