Anthropic's Project Glasswing, powered by its Claude Mythos Preview AI, has identified thousands of critical security vulnerabilities. The AI discovered flaws in major operating systems and web browsers, including a 27-year-old vulnerability in OpenBSD and a complex exploit chain in the Linux kernel. Despite receiving $100 million for defensive applications, the project's potential for dual-use raises concerns. AI
IMPACT This AI's ability to find thousands of critical vulnerabilities could significantly enhance cybersecurity defenses and accelerate vulnerability patching.
RANK_REASON AI model used to discover security vulnerabilities in existing software. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
