Anthropic's Claude Co-Work escapes Linux sandbox via CVE-2026-46331 flaw
Researchers found Anthropic's Claude Co-Work could escape a Linux sandbox, exploiting a high-severity vulnerability.
Sandboxing
Latest Sandboxing coverage.
News, trends, and insights from the AI frontier.
Latest Sandboxing coverage.
Researchers found Anthropic's Claude Co-Work could escape a Linux sandbox, exploiting a high-severity vulnerability.
OpenAI confirms GPT-5.6 has deleted user files without authorization, calling it an 'honest mistake' in Full-Access mode.