Skip to main content
🚀 Product launch⭐ Top story Verified88June 4, 2026

Anthropic Scales AI Vulnerability Hunter to 150 Partners Worldwide

Anthropic expanded Project Glasswing to 150+ organizations across 15+ countries, adding NATO, Samsung, and Okta.

Alibaba Opens Qwen to Third-Party AI Agents, Takes On ByteDanceSuno Hits $5.4B Valuation Despite Major Label Lawsuits
Insta's take

"Anthropic's AI is finding thousands of critical bugs that humans missed — and they're deliberately keeping it out of your hands. That gap between insiders and everyone else is growing fast."

Anthropic announced on June 2, 2026 that its restricted Mythos AI model — purpose-built for detecting software vulnerabilities — is now accessible to 150 additional organizations spanning more than 15 countries. The expansion triples the program's original footprint from its April launch with 50 partners, including the U.S. government. New entrants include Okta, Samsung, SK Hynix, SK Telecom, NATO, the EU's cybersecurity agency ENISA, and Rubrik, broadening the coalition well beyond its initial tech-heavy roster.

The June expansion deliberately targets underserved critical infrastructure sectors — power, water, healthcare, communications, and hardware manufacturing. Since launch, Project Glasswing partners have collectively uncovered more than 10,000 high or critical-severity security flaws, with several organizations reporting bug discovery rates increasing more than tenfold. Anthropic is backing the initiative with up to $100 million in Mythos Preview usage credits and $4 million in direct donations to open-source security organizations. Mythos itself will not be released to the general public due to misuse risk.

The move positions Anthropic as a dominant force in AI-powered enterprise security — particularly as OpenAI begins rolling out its own GPT-5.5-Cyber model to partners. With a confidential IPO filing and a valuation approaching $1 trillion, Anthropic is converting Glasswing into both a safety credential and a commercial moat.

Why Insta thinks this matters

If your organization operates in critical infrastructure, healthcare, or hardware, AI-assisted vulnerability detection is no longer a future consideration — it is active competitive reality. Partners are finding bugs ten times faster, which resets expectations for security team productivity and vendor SLAs. Executives should assess whether their security stack is keeping pace or falling behind organizations already inside programs like Glasswing.

Sources
TechCrunchReutersCNBCCyberScoopVentureBeatSiliconANGLECybersecurity DiveBNN BloombergFinancial Times

Relevant tools

Pi
Inflection's empathetic personal AI focused on supportive, c...
4.1
Insta's Weekly Digest — every Sunday

More AI news

Insta Tool Finder

Find the right AI tool for your business

Chat with Insta and get matched to the right tool in seconds.

Try Insta Tool Finder →