Kraken operator joins Anthropic's Project Glasswing for access to restricted Mythos 5 AI
Payward, the parent company of crypto exchange Kraken, has joined Anthropic's Project Glasswing and gained access to Claude Mythos 5, a restricted AI model designed to find and fix critical software vulnerabilities.
(Advertisement)
@krakenfx parent company @Payward has joined @AnthropicAI's Project Glasswing, an invite-only cybersecurity initiative that gives vetted organisations access to Claude Mythos 5, a restricted AI model built specifically to find and fix software vulnerabilities. The move marks one of the first crypto-native firms to enter the program.
What is Project Glasswing?
Project Glasswing launched on April 7, 2026, as Anthropic's structured approach to putting advanced AI in the hands of cybersecurity professionals. The initiative brings together major organisations including Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks. Anthropic has committed up to $100 million in usage credits and $4 million in donations to open-source security organisations to support the work.
Claude Mythos 5 is a restricted variant of Anthropic's frontier model, with certain safeguards lifted for vetted partners in cybersecurity and life sciences. It is not available through standard channels. Access requires an invitation through Project Glasswing, a signed NDA, and approval from Anthropic in consultation with the US government. Anthropic and its partners have used Claude Mythos Preview to find more than ten thousand high- or critical-severity vulnerabilities across the most systemically important software in the world.
What Payward Plans to Do
Access for Payward followed a US government decision allowing the model to reach critical-infrastructure operators. The firm said it will begin scanning all its environments for flaws in the coming weeks. Co-CEO Arjun Sethi framed the move in direct terms: "Defense has to compound as fast as offense."
Payward's security team plans to use the AI model not just for internal code review but also to report discovered vulnerabilities back to the maintainers of the software they depend on. The project places particular emphasis on open-source software, which forms the foundation of most modern systems, including critical infrastructure, yet whose maintainers have historically lacked access to sophisticated security resources.
Anthropic executives have indicated that without coordinated action, such tools could eventually reach actors who might deploy them for malicious purposes rather than defensive security work. For a major crypto exchange handling user funds at scale, the stakes of that arms race are especially high.
Sources:
Anthropic: Project Glasswing
Crypto Briefing: Payward joins Anthropic's Project Glasswing
CSO Online: Anthropic grants Project Glasswing access to 150 more companies
(Advertisement)
Latest News
Read More...
Author
Crypto RichRich has been researching cryptocurrency and blockchain technology for eight years and has served as a senior analyst at BSCN since its founding in 2020. He focuses on fundamental analysis of early-stage crypto projects and tokens and has published in-depth research reports on over 200 emerging protocols. Rich also writes about broader technology and scientific trends and maintains active involvement in the crypto community through X/Twitter Spaces, and leading industry events.