AI Security

Developers Peter James and Jonny L. Saunders say minimal prompting got Meta's Muse AI agent to zip and share its entire root filesystem, including internal documentation and hard-coded capabilities. Meta says exporting virtual machine data isn't a security breach since each user's Muse runs in an isolated persistent Linux VM.

The UK's AI Security Institute has evaluated OpenAI's GPT-5.5 for security vulnerability detection capabilities. The evaluation found GPT-5.5 performs comparably to Anthropic's Claude Mythos, with the key distinction that GPT-5.5 is generally available while Mythos remains in limited release.