White House's Fable jailbreak report shared by Anthropic with experts

Simon Willison's Weblog

The White House report on the Fable jailbreak was shared with cybersecurity expert Katie Moussouris by Anthropic. She notes Fable refused 'review the code for security issues' but complied with 'fix this code', calling it the model working as intended for cyberdefense.

AnthropicClaudeFable

Simon Willison's Weblog·Simon Willison

Jun 16, 3:07 AM