Kraken parent Payward has joined Anthropic’s Project Glasswing, according to CoinDesk, giving the crypto company access to Claude Mythos 5 for defensive cybersecurity work. CoinDesk reports that Project Glasswing is Anthropic’s initiative for allowing selected firms to use restricted frontier AI models to help secure critical software and infrastructure. The program already includes large technology companies including Amazon Web Services, Apple, Google, and Microsoft, as well as financial institutions including JPMorganChase, according to the report. Payward, which CoinDesk identifies as Wyoming-based, said Monday that it plans to deploy Mythos 5 across its environments in the coming weeks. The findings will feed into Payward’s existing security program, and the company also plans to disclose vulnerabilities found in third-party open-source software to the relevant project maintainers. Claude Mythos 5 is described in the report as Anthropic’s most advanced model for defensive cybersecurity. CoinDesk says the model is designed to analyze code at scale, identify vulnerabilities, and help developers fix them. Access has initially been limited to organizations that operate or defend critical infrastructure while Anthropic works on safeguards for wider availability. The timing matters because Payward is framing crypto infrastructure as a defensive target similar to other critical financial systems. CoinDesk reports that Payward argued crypto platforms face security challenges comparable to critical financial infrastructure because exchanges, custody systems, and settlement rails operate continuously and can be lucrative targets for attackers. The report also places the move in a broader security race: AI can help attackers find vulnerabilities faster, automate attacks, and make phishing or social-engineering campaigns more convincing. Payward co-CEO Arjun Sethi, quoted by CoinDesk, said security has historically favored attackers because they need only one flaw while defenders have to find all of them first; he argued that frontier AI can change that imbalance. According to Payward, the move follows the U.S. government making Mythos 5 available to U.S. organizations that operate and defend critical infrastructure. CoinDesk reports that Anthropic launched Project Glasswing in April 2026 and has since expanded it into technology and finance organizations, with Anthropic saying it plans to broaden access to Mythos-class cybersecurity capabilities as safeguards mature. Who benefits: Payward’s security team could benefit if Mythos 5 finds vulnerabilities earlier or helps developers fix them faster. Open-source maintainers may also benefit if Payward follows through on sharing third-party vulnerability findings. Who's exposed: Crypto platforms remain exposed to the same AI-assisted attacker tactics CoinDesk describes: faster vulnerability discovery, automated attacks, and more convincing phishing or social engineering. The report does not provide evidence yet on how effective Payward’s deployment will be in practice.