Payward Taps Anthropic’s AI for Defensive Cybersecurity in Project Glasswing — Payward, Inc.

Event summary

Payward, Inc. joins Anthropic’s Project Glasswing, gaining early access to Claude Mythos 5 for defensive cybersecurity.
The initiative aims to secure critical software by leveraging AI to identify vulnerabilities before attackers can exploit them.
Payward’s security team will integrate findings into its existing triage and remediation pipeline, including red and blue team exercises and a bug bounty program.
Discovered flaws in third-party open-source code will be reported upstream to project maintainers under responsible disclosure practices.
Project Glasswing launched in April 2026 and has already surfaced thousands of high or critical-severity security flaws.

The big picture

Payward’s participation in Project Glasswing reflects the growing arms race in AI-powered cybersecurity, where defenders now wield the same machine-scale vulnerability detection as potential attackers. As critical financial infrastructure becomes an increasingly attractive target, the ability to preemptively identify and patch flaws will be a key differentiator. Payward’s commitment to open-source contributions also underscores the industry’s shift toward collective defense mechanisms in an era of interconnected systems.

What we're watching

AI Defense Scaling: How quickly Payward can operationalize Mythos 5 findings to outpace emerging threats.
Ecosystem Hardening: Whether Payward’s open-source contributions will accelerate broader crypto ecosystem security.
Regulatory Alignment: The pace at which US policies adapt to AI-driven cybersecurity advancements in critical infrastructure.