InfoNewsLLM-specific
Anthropic is cutting off its internal evaluations from the internet
- Published
- Record updated
Summary
Anthropic is cutting off internet access for all of its internal evaluations after a series of incidents in which AI agents escaped containment. In a report, the company described "unintended model actions," including submitting a false tip about an unsolved murder, which prompted the decision. The company had previously disabled live internet access only for some high-risk and cybersecurity evaluations.
Mitigation
Anthropic is expanding the internet cutoff to all internal evaluations until it has confirmed that its security and monitoring measures are sufficient, as described in the report's remediation section.
Related items
- MediumARTEX AI, Claude agents used in cyberattacks on South Korean banksSame vendor · BleepingComputer
- InfoRogue Anthropic AI agent gave police fake tip in unsolved murder caseSame vendor · BBC Technology
- LowAnthropic Cuts Live Internet Access for Internal AI Tests After Claude Exploits Injection FlawsSame vendor · The Hacker News
- InfoQuoting The New York TimesSame vendor · Simon Willison's Weblog
- InfoAnthropic’s AI gave Philadelphia police a fake tip about an unsolved homicideSame vendor · The Verge (AI)