Expanding the Cyber Verification Program
This mainly matters if your organisation does hands-on cybersecurity defence or safety-testing work with AI models — it widens who can get verified access to Claude's more advanced cyber capabilities, under stricter monitoring and data-retention terms. For most business users this is not something you would use directly, but it signals Anthropic tightening both access and oversight around higher-risk AI capabilities.
What changed
Source summary
Anthropic expanded its Cyber Verification Program, adding tiered access (including a "Defense Access" tier for defensive cybersecurity work and a "Specialized Access" tier for testing critical safety systems) that gives verified security professionals deeper access to Claude's cyber capabilities, with verification requirements and security controls that scale with the tier. All participants must agree to data retention for misuse monitoring, though zero-retention options exist for certain model versions pending Anthropic's planned Enterprise Frontier Safeguards later in fall 2026.
Why it matters to your work
Practical interpretation
This mainly matters if your organisation does hands-on cybersecurity defence or safety-testing work with AI models — it widens who can get verified access to Claude's more advanced cyber capabilities, under stricter monitoring and data-retention terms. For most business users this is not something you would use directly, but it signals Anthropic tightening both access and oversight around higher-risk AI capabilities.
Source & availability
This is a summary of a published provider announcement, with practical context from SCALA’s feed.
Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.
www.anthropic.comThis update is primarily for technical teams.