Anthropic expands cyber access program to three tiers

Anthropic has relaunched its Cyber Verification Program (CVP) with three access tiers. Each tier includes Claude Opus 5.5, Claude Sonnet 5.5, Claude Mythos 5.1 and future models, with fewer blocking classifiers than the generally available versions. The company says its public models keep "conservative cyber safeguards that block most cyber work."

Until now, access ran through two separate programs. Project Glasswing gave Mythos to a small group of organizations securing critical software. The original CVP gave vetted security teams reduced safeguards on Opus and Sonnet.

• Defense Access covers SOC and incident response work, malware reverse-engineering, and vulnerability validation. It is open to companies, critical-infrastructure operators of any size, open-source maintainers and individual researchers with a track record of reported vulnerabilities. Anthropic aims to respond to applications within a few days.
• Red Team Access adds authorized penetration testing. It is for organizations only and reviews take a few weeks. Actions such as deploying ransomware or damaging physical systems are still blocked.
• Specialized Access has the fewest blocks. It is for organizations authorized to test systems like power grids, flight systems and interbank networks, and each one is reviewed with the US government. Existing Glasswing members move into this tier automatically.

Members must allow data retention so Anthropic can monitor for misuse. Its Enterprise Frontier Safeguards product, expected this fall, will let organizations keep that data in cloud infrastructure they control.

Anthropic tested the tiers by running Claude Opus 5.5 on CyScenarioBench, five attempts at each of 10 challenges. Without CVP access, every task was blocked on the first prompt. In Defense Access, 46 of 50 trials were blocked at some point. In Red Team Access, there were no blocks and the model completed 34 of 50 tasks. The source text ends before any Specialized Access results.