Anthropic Expands Claude Mythos to 15 Countries Targeting Critical Infrastructure

On Tuesday, AI company Anthropic announced a major expansion of its collaborative industry initiative Project Glasswing, extending the core AI model Claude Mythos to roughly 150 new organizations across more than 15 countries. The program aims to leverage AI for autonomous discovery and remediation of critical software vulnerabilities. This move marks a shift from government-focused collaboration to a broader rollout across global critical infrastructure.

According to the announcement, Anthropic had already provided an early preview of Claude Mythos to 50 initial partners—including the U.S. government—in early April for scanning codebases for security flaws. The new wave of organizations spans sectors such as energy, water, healthcare, telecommunications, and hardware, which were previously underrepresented. Publicly disclosed trusted organizations include U.S. identity security firm Okta, South Korean companies Samsung, SK Hynix, SK Telecom, NATO Headquarters, and the European Union Agency for Cybersecurity (ENISA).

Outlook & Analysis: Anthropic stated that it expects other AI companies to soon release models with capabilities comparable to Mythos Preview. In response, it is accelerating the implementation of safety guardrails through Project Glasswing. This initiative brings AI-powered vulnerability detection from government domains down to the core operations that sustain society, potentially ushering critical infrastructure protection into an "AI-native" phase.