Anthropic announced on October 6, 2026 the expansion of its Cyber Verification Program with three access tiers: Defense Access, Red Team Access, and Specialized Access. Qualifying security organizations can now use advanced Claude Opus 5.5, Sonnet 5.5, and Mythos 5.1 models for both defensive and offensive work, subject to tier-specific requirements. Defense Access covers defensive tasks like security operations center work and malware reverse engineering, while Red Team Access is limited to authorized organizations for penetration testing, excluding individual freelancers. Specialized Access allows testing of critical systems with prior US government review. CyScenarioBench evaluation results show that Defense Access blocked the majority of complex offensive scenarios, while Red Team Access successfully completed 34 out of 50 assessed tasks.
Source: Read the original article
