Grey Noise Launches AI Cyber Index, Benchmarking 46 Models on Real Defensive Security Work
Summary: GreyNoise launched the AI Cyber Index on Nov. 12, 2026, a benchmark built from eight years of real attack data that evaluates 46 closed and open-weight models on defensive security tasks. The Index reports per-task capability, refusal rates, and time and cost per problem alongside a composite leaderboard, drawing on labeled data used in production by more than 100,000 defenders.
Key Takeaway: GreyNoise states the time between vulnerability disclosure and exploitation has shrunk from days to minutes, and that edge systems such as firewalls, routers, and zero-trust proxies are now attackers' primary access vector - the same systems that cannot run endpoint agents and offer sparse telemetry.
Read the original at GreyNoise Blog ↗
- Topics
- Vulnerabilities & Patching
- Related
- Attackers Exploit UnrealIRCd Flaw That Enables Server-Level Command Injection Cyber Risks · Nov 14, 2026U.S. CISA adds Check Point and UnrealIRCd flaws to its Known Exploited Vulnerabilities catalog Cyber Risks · Nov 14, 2026PostgreSQL Patches 52 Flaws, Including 22-Year-Old Bug Enabling Authenticated RCE Cyber Risks · Nov 13, 2026