Claude sent a false homicide tip to police – and it’s part of a troubling AI pattern
ZDNET AIen
Models routinely take shortcuts to pass evaluations. Who gets to decide what incidents are innocuous?
This is a short summary published by AI Global Wire. The full article is owned and hosted by ZDNET AI — open it there to read it in full.
Read the full story at ZDNET AI- Anthropic
- Företag
Related AI news
- UK given access to safety-test latest AI models, minister saysEconomic Times Tech · October 12, 2026
- Meta releases Muse Wonder AI models to challenge Anthropic and OpenAIEconomic Times Tech · October 12, 2026
- Anthropic's Claude AI submitted false murder tip to Philadelphia policeArs Technica AI · October 12, 2026
- A profile of Microsoft EVP of Copilot Charles Lamanna, a potential Satya Nadella successor; a source says Microsoft is on track to pay Anthropic $2B+ per year (Aaron Holmes/The Information)Techmeme · October 12, 2026
- White House reverses course and restores UK AI Safety Institute's access to frontier modelsThe Decoder · October 12, 2026
- Mistlabs takes its Ambi AI system out of beta on Apple devices with $26M in fundingSiliconANGLE · October 12, 2026