Expanding Project Glasswing

2026-06-01 · Source: Anthropic News · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Intermediate, medium

Summary

Anthropic is significantly expanding Project Glasswing, its collaborative initiative to secure critical software globally. Initially launched in early April with approximately 50 partners, the program utilized Claude Mythos Preview to scan codebases, identifying over 10,000 high- or critical-severity security flaws. The expansion now includes around 150 new organizations across more than 15 countries, encompassing vital sectors like power, water, healthcare, communications, and hardware, many of which are vendors maintaining codebases relied upon by governments and other entities. This initiative addresses the imminent threat of cheap, fast AI models with powerful cyber capabilities, which Anthropic expects other companies to release within 6-12 months, potentially without safeguards. Project Glasswing aims to help the software industry adapt by providing safe access to advanced models and tools, shifting focus from vulnerability discovery to disclosure, patching, and deployment of secure software.

Key takeaway

For Security Engineers adapting to the rapid evolution of AI-driven cyber threats, you must proactively integrate advanced AI models into your defense strategies. Anthropic's Project Glasswing demonstrates that tools like Claude Mythos Preview can identify and help patch critical vulnerabilities at scale. Begin exploring AI-powered code scanning, penetration testing, and automated threat response to maintain pace with emerging threats and secure your organization's critical infrastructure.

Key insights

AI models are rapidly changing cybersecurity, necessitating industry-wide adaptation and robust defensive tools.

Principles

• AI will make all software more secure.
• Cyberdefenders must adapt to AI-driven threats.
• Safeguards are crucial for powerful AI cyber models.

Method

Organizations deploy Claude Mythos Preview to scan codebases for vulnerabilities, share findings, and then use the model for patching, pre-release checks, and other defensive tasks.

In practice

• Utilize Claude Security for code scanning.
• Employ AI models for penetration testing.
• Automate threat detection and response.

Topics

• Project Glasswing
• Claude Mythos Preview
• Cybersecurity
• Vulnerability Management
• AI Security
• Critical Infrastructure

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Security Engineer, Security Engineer, Policy Maker

Related on AIssential

• Project Glasswing is World’s Most Powerful AI in Action — Anthropic's Mythos Preview AI model significantly advances defensive cybersecurity by autonomously finding and exploiting complex vulnerabilities.
• Anthropic's Project Glasswing - restricting Claude Mythos to security researchers - sounds necessary to me — Advanced LLMs are now autonomously finding and exploiting critical software vulnerabilities at an unprecedented scale.
• Anthropic scales Project Glasswing to 150 partners across 15 countries to hunt critical software flaws — AI models like Claude Mythos can rapidly identify critical software vulnerabilities, posing both defensive and offensive capabilities.
• Anthropic is launching a new AI model for cybersecurity - The Verge — Anthropic's Claude Mythos Preview autonomously identifies and exploits system vulnerabilities, enhancing cybersecurity defenses.
• Anthropic, Big Tech launch Project Glasswing for AI-led cybersecurity - Business Standard — Project Glasswing deploys Claude Mythos AI for defensive cybersecurity, identifying vulnerabilities across major systems.
• Verizon Joins Anthropic’s Project Glasswing for AI Defence — Advanced AI models like Claude Mythos Preview can proactively identify critical zero-day vulnerabilities, necessitating collaborative defense initiatives like Project Glasswing.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Anthropic News.