securityweek.comarticleAnthropic Introduces 3-Tier Cyber Verification Program for AI Access
Anthropic's revamped Cyber Verification Program merges the original CVP with Project Glasswing into three tiers: Defense Access, Red Team Access and Specialized Access. All tiers cover Opus 5.5, Sonnet 5.5, Mythos 5.1 and future models, with different verification and fewer cyber blocks at higher tiers. Applicants must accept data retention, and availability differs across the Claude Platform, Vertex AI, Microsoft Foundry and Amazon Bedrock.
TL;DR Anthropic merges its Cyber Verification Program and Project Glasswing into three access tiers that relax cyber safeguards on its top models for vetted users.
Also seen from another angle🛡️ Security →
Organisations named here
- Anthropic
- Google Cloud
- Microsoft
- Amazon
- U.S. government
- AWS
Technologies named here
- Claude Opus 5.5
- Claude Platform
- Claude Sonnet 5.5
- Microsoft Foundry
- Claude Fable 5.1
- Claude Mythos 5.1
- Amazon Bedrock
- Claude Mythos
- Cyber Verification Program
- Claude
- Enterprise Frontier Safeguards
- Project Glasswing
- Vertex AI
Concepts named here
- Cyber safeguards
- Cyber Verification Program
- Enterprise Frontier Safeguards
- Project Glasswing
- Tiered access verification
- Zero data retention
Related in ai-policy
futureoflife.orgarticleFuture of Life Institute AI Safety Index — Summer 2026
interconnects.aiarticleLessons from the hacks
interconnects.aiarticle