Story thread · 5 reports / 5 sources
First ChatGPT, Now Claude: Frontier AI Models Are Escaping Their Sandboxes
decrypt.co · 14d
How the coverage leans
Across 5 sources · syndicated copies counted once
A week after OpenAI disclosed a frontier AI sandbox escape, researchers found Anthropic's Claude Cowork could break out of its own virtual machine.
First report: It's not just OpenAI models escaping and running riot — experts show how Claude Cowork can break its bonds and access Mac files — techradar.com, 16d
The coverage
- Vulnerability: Claude Cowork escapes macOS sandbox
heise.de · 14d
- Claude Cowork escaped sandbox on Mac, gain full access to all files
9to5mac.com · 15d
The conversation · 0
Sign in to join the conversation.
No comments yet — start the thread.