The government believes it has become aware of a method of bypassing, or 'jailbreaking' Fable 5.

in 89% of the 198 manually reviewed vulnerability reports, our expert contractors agreed with Claude's severity assessment exactly, and 98% of the assessments were within one severity level. If these results hold consistently for our remaining findings, we would have over a thousand more critical severity vulnerabilities and thousands more high severity vulnerabilities.

This attack achieved a high success rate against state-of-the-art models, including Claude Opus 4.7.

The most urgent finding this week comes from researchers who demonstrated that the very mechanism enabling agents to use tools - function calling - can be hijacked with alarming reliability.

just a handful of obviously fake articles could cause Gemini, ChatGPT, and Copilot to inform users about an imaginary disease with a ridiculous name.

select known-vulnerable dependency versions 50% more often than humans.

current systems remain highly vulnerable

current systems remain highly vulnerable