Claude Cowork Flaw Could Let AI Agent Escape Its VM and Access Mac Files

Cybersecurity researchers discovered a sandbox escape vulnerability in Anthropic's Claude Cowork that allows AI agents running in a Linux VM to access files anywhere on a host Mac, affecting ~500,000 macOS users before patching. Anthropic closed the report without issuing a fix, instead defaulting new deployments to cloud execution, leaving local-run users still exposed.

This Wire brief sits within Fusion42's coverage of Cybersecurity. Wire is Fusion42's founder-focused intelligence feed: each story is connected to the funds and startups it names — every one with a live profile on Raise or Scout — so founders can follow the capital and the momentum behind the headline rather than just the headline itself. Wire analysis is one of the live surfaces Arthur reasons over.