Payward Joins Anthropic’s Project Glasswing, Deploying Claude Mythos 5 to Secure Financial Infrastructure
Payward, the parent company of cryptocurrency exchange Kraken, has joined Anthropic’s Project Glasswing. According to Yahoo Tech and Crypto Briefing coverage, the company is deploying the Claude Mythos 5 model to scan its environments for software vulnerabilities and fortify its around-the-clock financial infrastructure against automated cyber threats.
The Bottom Line
- Defensive Scaling: Payward becomes the first reported crypto company to gain access to Anthropic’s Claude Mythos 5.
- Asymmetry Shift: The integration addresses machine-scale threats by allowing automated code analysis to match sophisticated attacker capabilities.
- Ecosystem Impact: Discovered third-party open-source flaws will be reported to project maintainers, extending protection across the crypto industry.
Bridging the Asymmetry Gap in Digital Asset Defense
Financial technology infrastructure operates under relentless pressure. Exchanges, custody systems, and settlement rails run continuously, creating an environment where defenders face an uphill battle against automated exploits. Crypto platforms face security challenges similar to other critical financial infrastructure.
Here is the math: attackers only need to locate a single overlooked weakness in a codebase, whereas defenders must identify and patch every single vulnerability first, every single day. Project Glasswing was initiated by Anthropic in April 2026—as reported by Crypto Briefing—after the organization discovered that sophisticated artificial intelligence systems could spot and exploit code flaws better than nearly all human experts. By incorporating Claude Mythos 5 into its operations, Payward aims to bridge that capability gap.
Inside the Claude Mythos 5 Integration
Claude Mythos 5 is Anthropic’s most advanced model for defensive cybersecurity. The system is designed to analyze code at scale to identify vulnerabilities before malicious actors can weaponize them. Yahoo Tech notes that Mozilla previously reported Claude Mythos identified 271 vulnerabilities in Firefox during testing, demonstrating the model’s analytical throughput.
Payward’s security division—which already maintains ISO 27001 and SOC 2 certifications alongside dedicated red and blue teams—will channel Mythos 5 findings into its existing triage and remediation process. Beyond internal asset protection, the firm announced that any flaws discovered in third-party open-source software will be reported to respective project maintainers. This disclosure model ensures that fixes propagate outward, hardening the broader crypto ecosystem.
Collective Security Demands Across the Crypto Sector
This deployment arrives on the heels of a coordinated push by industry participants. Last week, an open letter organized by the Bitcoin Policy Institute and signed by over 40 Bitcoin and cryptocurrency companies—including Ark Invest, Coinbase, Block, and BitGo—urged major AI labs to open trusted-access programs to qualified defenders.
The urgency behind these initiatives stems from a fundamental shift in threat actor capabilities. Cryptocurrency enterprises face an escalating hazard as malicious operators leverage artificial intelligence to accelerate weakness detection, execute automated assaults, and craft more persuasive phishing and social-engineering scams.
| Metric / Parameter | Reported Data | Source Reference |
|---|---|---|
| Participating Crypto Signatories to AI Letter | 40+ companies | Yahoo Tech / Crypto Briefing |
| Primary Defensive AI Model Deployed | Claude Mythos 5 | Payward Release |
| Vulnerability Severity Focus | High and Critical | Project Glasswing Data |
The Strategic Realignment of Financial Security
“Security has always been an unfair game. An attacker needs to find one flaw. A defender has to find all of them, first, every single day,” Payward Co-CEO Arjun Sethi stated in the release. “Frontier AI is the first thing that flips that asymmetry. A model can read every line of code the way an attacker would, at machine scale, so we find the flaw before anyone can build the exploit.”
As Project Glasswing expands its footprint, the inclusion of a crypto company signals a maturation in how digital asset exchanges manage operational risk. As Anthropic continues developing safeguards for a wider rollout, access to Mythos-class cybersecurity capabilities remains limited to organizations that operate or defend critical infrastructure.
Disclaimer: The information provided in this article is for educational and informational purposes only and does not constitute financial advice.
