[AINews] Gemini 4 Carbon tops leaderboards, launches for cyber defenders

⚡ Two weeks after Argon, Google has a new No. 1 AI model Gemini 4 Carbon is Google's next frontier model, and it tops the Artificial Analysis Intelligence Index with 59 points. Claude Opus 5.5 has 58, GPT-6 Astra 57. The biggest jump is in cybersecurity. Carbon scores 88.2% on SEC-bench Pro, a test of finding vulnerabilities. Google's Chrome team says it did over 10 times better than another lab's large frontier model at writing high-quality security patches. It isn't first everywhere. On coding agents it ranks second, 59 points to 62 for Claude Code running Opus 5.5. You can't use it yet. Some cyber defenders get access this week, then trusted partners, with "wider availability soon." The introductory price is $2 per million input tokens and $10 per million output, later $4 and $20. @aipost 🏴

Google announces Gemini 4 Carbon

Gemini 4 Carbon: Google's Best AI and You CAN'T Use It

Google announces Gemini 4 Carbon, its latest frontier AI model that you can't use. Via @arstechnica #AI #ArtificialIntelligence 💻 🤖 🧠 Google announces Gemini 4 Carb...

Google announced Gemini 4 Carbon on Oct 28, calling it its strongest Gemini 4-era model, built for cybersecurity and long-horizon engineering work. Per Google’s figures, it scores 88.2% on SEC-bench Pro (finding real vulnerabilities), 54.8% on ExploitGym (validating exploits), and 71.0% on CWE-bench (patching), and it tops the Artificial Analysis Intelligence Index v5 at 59. These are company-reported numbers; independent reproduction is still pending. The more interesting part is the rollout: Carbon goes first to a subset of vetted cyber defenders through Google’s Fairwind Program, with developers, businesses and consumers to follow later (no date given). Google says it is also taking part in the U.S. government’s voluntary pre-release testing process. Discussion: a model this good at fin