Anthropic announced Tuesday that it will expand Project Glasswing, a collaborative industry initiative to use AI to discover and fix critical software vulnerabilities, to approximately 150 new organizations in more than 15 countries.
The news comes a day after Anthropic announced it had secretly filed for an initial public offering after a $65 billion funding round at a valuation of about $1 trillion.
Anthropic’s Claude Mythos is at the heart of Project Glasswing. The AI company called the model its most powerful yet, able to identify thousands of zero-day vulnerabilities over several weeks. In early April, Anthropic granted 50 early partners, including the U.S. government, access to Claude Mythos Preview to scan their codebase for vulnerabilities and security flaws.
The growing list of organizations that currently have access to Mythos includes power, water, healthcare, communications and hardware, industries that the company said were “under-represented” in Anthropic’s first cohort. Anthropic noted in a blog post that many of those who will gain access will be companies and nonprofits that manage codebases that other organizations and governments rely on.
“What each of our partners has in common is that a successful attack on their codebase can have devastating effects,” the company said. “For most partners, we estimate that a large-scale attack could affect more than 100 million people and have significant implications for both global and national security.”
The expanded group includes organizations from countries friendly to the United States, including Australia, Canada, France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, Sweden, India, Japan, New Zealand and South Korea, the Financial Times reported, citing people familiar with the matter.
The FT also reported on several organizations that were given access to Mythos. These include US-based identity and security management tool Okta, Korean companies Samsung, SK Hynix, and SK Telecom. NATO, a US-led military alliance headquartered in Brussels. EU cybersecurity agency ENISA.
TechCrunch has reached out to Anthropic for confirmation.
Anthropic says it expects other AI companies to soon develop models with similar capabilities to Mythos Preview, which is why the company is rushing to establish safeguards within Project Glasswing.
Since the release of Mythos, rival OpenAI has released its own cybersecurity-focused model, GPT-5.5-Cyber, and deployed it to a large group of partners for testing.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.
Source link
