NewsThe Hacker News29.10.2026

Google's Gemini 4 Carbon Leads Benchmarks for Finding, Exploiting, and Patching Vulnerabilities

Google on Wednesday took the wraps off Gemini 4 Carbon, its latest frontier model with state-of-the-art cybersecurity and long-horizon capabilities. "It shows significant gains across coding and agentic tasks as well as across key capabilities critical for cybersecurity defense, landing a firm lead in SEC-bench Pro (88.2%) for finding vulnerabilities, ExploitGym (54.8%) for exploitation, and

Other advisories

Not ratedosv:CVE-2026-29075Mesa: Checking out of untrusted code in `benchmarks.yml` workflow may lead to code execution in privileged runner
Not ratedosv:MAL-2025-1146Malicious code in lighthousebenchmarks (npm)
Not ratedosv:MAL-2024-1051Malicious code in build-benchmarks (npm)
Not ratedosv:MAL-2023-8453Malicious code in classnames-benchmarks (npm)
Not ratedosv:MAL-2022-4485Malicious code in marketplace-benchmarks (npm)
Not ratedosv:MAL-2022-5278Malicious code in perf-benchmarks-components (npm)
Not ratedosv:MAL-2022-1865Malicious code in cfbenchmarks-adapter (npm)

Other news entries

So Many AI Security Benchmarks, but None Tell the Whole StoryNewsDark Reading27.10.2026
On AI SecurityNewsSchneier on Security
Why AI Needs a “Genie Coefficient”NewsSchneier on Security
Googles Gemini 4 Argon legt Fokus auf CybersicherheitNewsheise Security
Claude Fable 5.1 setzt sich in Benchmarks an die Spitze, bleibt aber teuerNewsheise Security

Source: https://thehackernews.com/2026/10/googles-gemini-4-carbon-leads.html

More on Benchmarks

Other advisories

Other news entries