wired.comOne of China’s Most Powerful AI Models Has Also Escaped ContainmentSecurity researchers say that Kimi K3, an open-weight model from China, wandered off to the internet in an attempt to cheat on a test it was given.ShareSave1mo ago
wired.comAnthropic Says Claude Hacked Real Systems During Cybersecurity TestsAnthropic discovered three of its AI models had breached real organizations during third-party evaluations.ShareSave1mo ago
