Anthropic Expands Access to Advanced AI Models for Cybersecurity Testing
Anthropic is expanding its Cyber Verification Program (CVP) to allow more vetted cybersecurity professionals to test its most advanced AI models with fewer safeguards. This follows its Project Glasswing initiative, which uncovered over 134,000 software vulnerabilities between April and October, including more than 33,000 rated critical or high severity. The revamped CVP combines previous programs and offers three tiers of access with varying verification and security controls, aiming to enhance software security through collaboration with security teams, infrastructure operators, and researchers.
First-hand measurement across 3 sources
We measured how 3 outlets covered this story. No outlet gave this story a measurable political slant — there is no left–right reading to report. Overall sentiment is neutral (59/100). Lens Score 43/100.
Outlets measured: firstpost, thehindu, economictimes. See how each one headlined and framed the same story in the source comparison below.
AI Analysis
Sentiment was consistent across outlets (58–62/100), indicating broadly factual reporting rather than editorialising.
Coverage timeline
economictimes broke this story on 6 Oct, 07:10 pm. Other outlets followed.