Anthropic is significantly broadening the reach of Project Glasswing, an industry-focused initiative designed to identify and remediate critical software vulnerabilities using advanced artificial intelligence. The program is now being extended to approximately 150 new organizations spanning more than 15 countries, marking a major escalation in the effort to secure essential digital systems.

At the core of this initiative is Claude Mythos, an AI model specifically engineered to detect thousands of zero-day vulnerabilities within complex codebases. Following a successful pilot phase with 50 initial partners—including various U.S. government entities—the expanded rollout targets sectors deemed vital to global stability, such as power grids, water management, healthcare, communications, and hardware manufacturing. Anthropic emphasized that the new cohort consists of organizations whose codebases serve as foundational infrastructure for other entities and governments worldwide.

The expansion includes key international partners across Australia, Canada, Japan, South Korea, India, and several European nations. Notable participants now utilizing the technology include major industry players like Samsung, SK Hynix, and Okta, as well as institutional bodies such as NATO and the European Union’s cybersecurity agency, ENISA. Anthropic has underscored the urgency of this deployment, noting that a successful cyberattack on these specific codebases could potentially impact over 100 million people, posing significant risks to national and global security.

As the competitive landscape for cybersecurity AI intensifies, with rivals like OpenAI introducing specialized models such as GPT-5.5-Cyber, Anthropic is prioritizing the rapid establishment of safety protocols. By integrating Claude Mythos into the workflows of these critical organizations, the company aims to create a robust defensive layer against emerging digital threats before they can be exploited by malicious actors.

Key Takeaways

• Anthropic is expanding its Project Glasswing initiative to 150 organizations across 15+ countries to bolster critical infrastructure security.
• The Claude Mythos AI model is being deployed to identify zero-day vulnerabilities in sectors including energy, healthcare, and communications.
• High-profile partners now include NATO, Samsung, and the EU’s cybersecurity agency, reflecting the global scale of the security effort.

Editor’s Analysis & Impact

The expansion of Project Glasswing signals a pivotal shift in how critical infrastructure is defended against increasingly sophisticated cyber threats. By leveraging AI models like Claude Mythos to proactively hunt for zero-day vulnerabilities, Anthropic is moving beyond traditional reactive security measures. This initiative highlights a growing trend where AI companies are positioning themselves as essential partners in national and global security frameworks. The inclusion of major international entities like NATO and ENISA suggests that AI-driven vulnerability management is becoming a standard requirement for maintaining digital sovereignty. As competition heats up with rivals like OpenAI, the race to establish these security safeguards will likely define the next era of enterprise software, forcing a paradigm shift where code integrity is continuously verified by autonomous systems rather than manual audits alone.

Frequently Asked Questions

Q: What is the primary goal of Project Glasswing?
A: Project Glasswing is an industry initiative aimed at identifying and fixing critical software vulnerabilities in essential infrastructure using Anthropic's AI model, Claude Mythos.

Q: Which industries are being prioritized in this expansion?
A: The expansion focuses on critical sectors including power, water, healthcare, communications, and hardware, where a security breach could have widespread, catastrophic consequences.