Exclusive: Anthropic’s Claude 3.7 sonnet is the safest AI model so far, the independent test finds

The new independent exploration of the Holistic AI, the British companies testing AI models, suggests that the new Anthropic model Claude 3.7 Sonnet AI cannot convince that it jumps its built-in protective fences, making it the safest AI model that has been released.

Photographic Illustration Cheng Xin – Getty Images