Anthropic Scales Claude Mythos to Critical Infrastructure in 15+ Countries

Anthropic Scales Claude Mythos to Critical Infrastructure in 15+ Countries">

Anthropic is significantly expanding Project Glasswing, its collaborative effort to identify and rectify critical software vulnerabilities using artificial intelligence. The initiative is being opened up to approximately 150 new organizations across more than 15 countries, the company announced on Tuesday. This development comes on the heels of Anthropic's confidential filing for an initial public offering, following a substantial $65 billion funding round that valued the company at nearly $1 trillion.

At the core of Project Glasswing is Anthropic's Claude Mythos, described by the firm as its most powerful model yet. Claude Mythos has the capability to detect thousands of zero-day vulnerabilities over several weeks. Earlier in April, Anthropic provided 50 initial partners, including the U.S.

government, with access to Claude Mythos Preview. This allowed them to scan their codebases for vulnerabilities and security flaws. The list of organizations now with access to Mythos encompasses a broad range of critical sectors including power, water, healthcare, communications, and hardware.

These are industries that were not adequately represented in Anthropic's initial cohort. Many of the new organizations maintain codebases that other organizations and governments rely on. According to Anthropic, what these partners have in common is the potential for a catastrophic impact if their codebase is successfully attacked.

The company estimates that a major attack could affect more than 100 million people, with significant ramifications for both global and national security. The expanded group includes organizations from countries closely aligned with the U.S., such as Australia, Canada, France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, Sweden, India, Japan, New Zealand, and South Korea. Some of the organizations that have been granted access to Mythos include Okta, a U.S.-based identity and security management tool; South Korean companies Samsung, SK Hynix, and SK Telecom; NATO; and the EU's cybersecurity agency ENISA.

Anthropic is racing to establish safeguards within Project Glasswing, as it expects other AI companies to soon develop models as capable as Mythos Preview. The company is aware that rival OpenAI has released its own cybersecurity-focused model, GPT-5.5-Cyber, which it has rolled out to a large group of partners for testing.