Breaking

Anthropic Expands Cyber Verification Program With Access to Mythos 5.1 and Latest Claude Models

October 6, 2026 at 15:02 EDT

Foundation Models
Security

Anthropic has announced plans to expand its Cyber Verification Program, granting verified security professionals access to Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5 with specially configured safeguards for defensive security work. The expansion represents a significant broadening of Anthropic's dual-use AI access strategy, moving beyond the current Opus and Sonnet models to include the company's most specialized cybersecurity-focused offering.

October 2026 • Anthropic

Advanced AI Security Tools for Verified Defenders

Anthropic expands the Cyber Verification Program with Mythos 5.1, Opus 5.5, and Sonnet 5.5 for defensive security professionals.

83.1%

Mythos cybersecurity benchmark

181×

Exploit success rate on Firefox

60%

Sonnet 5.5 lower token cost

Three access tiers: Defense Access for researchers, Red Team Access for organizations, and Specialized Access for advanced work. All require identity verification and phishing-resistant MFA (mandatory by December 15, 2026). Organizations already verified include Ridge Security, Cymulate, Strobes, Forescout, MIND, and Secure.com.

Restrictions remain: Mass data exfiltration, ransomware development, command-and-control infrastructure, and binary-based vulnerability scanning are prohibited regardless of verification status.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.