Anthropic launches Project Glasswing for critical software security
Anthropic announced Project Glasswing, a security initiative powered by its new Claude Mythos Preview model to find software vulnerabilities.
Anthropic announced Project Glasswing, an initiative aimed at securing critical software infrastructure. The project is powered by Claude Mythos Preview, a new frontier model that Anthropic says can detect software vulnerabilities at a level approaching the most skilled human security researchers.
The move signals a growing focus among AI labs on applying large models to cybersecurity — an area where the complexity and volume of code have long outpaced manual review capacity. By fielding a model specifically tuned for vulnerability discovery, Anthropic is positioning itself in the emerging market for AI-assisted security tooling, alongside efforts from competitors targeting code analysis and threat detection. Project Glasswing's emphasis on "the world's most critical software" suggests the initial focus will be on high-stakes infrastructure rather than general-purpose code review.