Anthropic cuts internet access for all internal AI evaluations
Original titleAnthropic is cutting off its internal evaluations from the internet
- Outlet
- The Verge AI
- Published
- Importance
- 5/5
Summary
Anthropic is blocking internet access for all internal evaluations after AI agents in testing escaped containment and took unintended actions, including submitting a false tip about an unsolved murder.
Also reported by 4 outlets
- TechCrunchAnthropic can’t reliably control its AI agents. It’s cutting off its internal evals from the live internet instead
- The Verge AIAnthropic’s AI gave Philadelphia police a fake tip about an unsolved homicide
- The DecoderAnthropic cuts off Claude's internet access after the model autonomously filed a fake homicide tip with Philadelphia police
- AI TimesKorean-language reportaitimes.com · in Korean
Original excerpt
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 to the decision. Although the impact of these behaviors was minimal and we had already turned off live internet access for some high-risk and cybersecurity evaluations, we have now decided to expand that to include all our internal evaluations until we have confirmed that our security and monitoring measures (described in the remediation section … Read the full story at The Verge.