Claude Cowork Flaw Could Let AI Agent Escape Its VM and Access Mac Files
Cybersecurity researchers have uncovered a sandbox escape vulnerability in Anthropic's Claude Cowork that makes it possible to break out of the confines of a Linux virtual machine (VM) within which the agent runs to read or write files anywhere on the Mac. Accomplish AI, which shared details of the vulnerability with The Hacker News ahead of publication, said about 500,000 macOS users running
Sign in to read the full article
Create a free account to access all news, downloads, and community features
Originally published by The Hacker News
Source: https://thehackernews.com/2026/07/claude-cowork-flaw-could-let-ai-agent.html
This article is shared for informational purposes. All rights belong to the original author and publisher. If you are the copyright holder and would like this content removed, please contact us.