Get the latest tech news
Anthropic reportedly withholds access to Mythos 5.1 from UK safety testing body
The decision marks the first time the AI Security Institute has been left out of pre-release evaluations of Anthropic models
None
Or read this on r/technology