Anthropic scales Claude Mythos to critical infrastructure in 15+ countries

Anthropic has announced a significant expansion of its AI security initiatives, scaling Claude Mythos access to critical infrastructure operators across 15 countries. The expansion represents a major step in deploying advanced AI capabilities to organizations responsible for essential services affecting hundreds of millions of people globally. Through Project Glasswing, its security vulnerability program, Anthropic is providing enhanced access to 150 organizations operating in power, water, healthcare, and communications sectors.

Anthropic's initiative focuses on critical infrastructure sectors where cybersecurity failures could have catastrophic consequences. The company is extending Claude Mythos—an advanced version of its Claude AI model—to organizations in 15 countries, enabling them to leverage AI capabilities for vulnerability detection, threat analysis, and security enhancement. Project Glasswing, the underlying security program, aims to identify and remediate vulnerabilities before malicious actors can exploit them. This coordinated approach combines AI access with structured security protocols to protect systems that collectively serve over 100 million people.

The expansion prioritizes sectors essential to modern society: electrical power grids, water treatment and distribution systems, healthcare networks, and telecommunications infrastructure. These sectors represent critical dependencies where disruption could trigger cascading failures across economic and social systems.

• Enhanced AI-driven security capabilities reduce vulnerability detection timelines and improve threat response coordination across critical sectors
• Standardized access to Claude Mythos creates consistency in security practices across geographic and organizational boundaries
• Integration of advanced AI into critical infrastructure operations establishes new standards for cybersecurity resilience in essential services
• Coordinated vulnerability disclosure through Project Glasswing strengthens the security ecosystem by enabling collaborative threat identification
• Expansion validates commercial AI models as viable tools for protecting critical national and international infrastructure

Anthropic's scaled deployment of Claude Mythos to critical infrastructure represents a pivotal moment in AI's role within cybersecurity. As cyberattacks against essential services increase in sophistication, deploying advanced AI capabilities to frontline defenders enhances collective security posture. This initiative demonstrates how private AI companies can contribute meaningfully to national and international security infrastructure while establishing frameworks for responsible AI deployment in high-stakes environments.