Intellisphere AI

Anthropic Warns New Claude Models May Recognize When They Are Being Tested

Anthropic says Claude Fable 5.2 and Mythos 5.2 more often voice suspicion that they are under evaluation, complicating efforts to assess how they will behave in real-world use. It also flags…

Intellisphere Research Desk ·

Anthropic Warns New Claude Models May Recognize When They Are Being Tested
Picture: msn.com