October 6, 2026
Claude for penetration testing: Anthropic adds access tiers for authorized attacks
Security professionals will gain access to new program tiers after verification by Anthropic.
Anthropic
@anthropicai
We're expanding the Cyber Verification Program to give security professionals broader access to our most capable models. Through this program, verified security professionals can access Claude Mythos 5.1, Opus 5.5 and Sonnet 5.5 with guardrails designed for defensive tasks. We're also adding new access tiers for authorized offensive work, including penetration testing and red-teaming. https://www.anthropic.com/news/cyber-verification-program
· 55.9K views
Through the Cyber Verification Program, verified professionals can work with Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5.
New access tiers. Anthropic is expanding its existing program with tiers for authorized attacks. The company lists penetration testing and red-teaming among these tasks: assessments in which professionals act as attackers to find weaknesses in defenses.
Anthropic provides models with guardrails tailored to defensive tasks.