Anthropic and Claude help Apple patch critical macOS vulnerabilities
Apple’s macOS Tahoe update fixes dozens of vulnerabilities, including critical flaws that allowed root access. The CVE-2026-28952 patch, discovered with help from Anthropic and Claude researchers, is among them.
The collaboration continues the Project Glasswing effort, which uses Claude Mythos Preview to find code bugs. This model previously uncovered thousands of vulnerabilities in open-source software like Firefox and Symfony.
AI-assisted security auditing is becoming standard for major vendors. Automated error detection finds complex logic and memory issues much faster than manual analysis.
Up next
Anthropic plans public release of Mythos models
Anthropic confirms plans to release Mythos, its vulnerability-discovery model, to the public. Currently, it’s restricted due to misuse risks. Mythos already finds thousands of critical code flaws, but Anthropic is still working on safeguards to prevent exploitation.
Related stories
- Claude Mythos discovers over 10,000 critical vulnerabilities
- Claude Mythos uncovers 19 vulnerabilities in Symfony
- Apiiro finds Claude Code secrets in 1,122 GitHub files
- Email leak stopgap in Claude Code checks only shell calls
- Claude Code agent sat on a Pentagon network for two months
- Irregular CEO blames human error for AI agent breaches
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.