# Google announces Gemini 4 Carbon, citing a major cybersecurity leap

- Published: 2026-10-28T17:00:58.000Z
- Updated: 2026-10-28T22:03:03.766Z
- Section: Models
- Topics: Research, Security
- Page: https://aidailyjournal.com/en/n/2026/10/28/google-announces-gemini-4-carbon-citing-a-major-cybersecurity-leap/
- Sources: [Google DeepMind](https://deepmind.google/blog/gemini-4-carbon-our-next-frontier-model), [The Verge](https://theverge.com/tech/1021339/google-gemini-4-carbon-cybersecurity-frontier), [Ars Technica](https://arstechnica.com/ai/2026/10/google-announces-gemini-4-carbon-its-latest-frontier-ai-model-that-you-cant-use)

## Summary

Google announced Gemini 4 Carbon, the successor to Gemini 4 Argon, saying the model represents a significant advance in cybersecurity, coding, and long-horizon agentic tasks. The company says it leads the Artificial Analysis Intelligence Index v5.0 and scores 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. Google also says the Chrome team saw performance more than ten times better than another lab’s frontier model when generating high-quality patches.

Access starts this week for a group of cyber defenders and will expand to trusted partners through the Fairwind Program before reaching developers, businesses, and consumers. The introductory price will be $2 per million input tokens and $10 per million output tokens. Ars Technica notes that Carbon beats Argon by narrow margins on the disclosed tests and questions whether it is an entirely new model or a post-training upgrade.

## Why it matters (editorial interpretation)

The staged rollout currently limits independent evaluation by developers and businesses. Because Ars Technica questions whether the advance over Argon is a new model or a post-training upgrade, Google’s reported results should be interpreted in that context.

_Summary written by AI Daily Journal in its own words, based on the sources above. For details, read the original articles._