THE CRUNCH

Anthropic has released a report alleging that China-based AI companies are mounting persistent distillation attacks against its models. The firm claims these campaigns have intensified in recent months as competition in the AI sector heats up. The attacks involve extracting knowledge from Anthropic's systems, a technique that could undermine the safety and proprietary nature of the underlying models.

WHAT HAPPENS NEXT

It remains unclear how Anthropic plans to mitigate these attacks or whether other AI labs are facing similar threats.