Google has unveiled Gemini 4 Carbon, its latest AI model with improved coding, reasoning and cybersecurity capabilities. The company says the model delivers a significant improvement in finding software vulnerabilities, testing potential attacks and fixing security flaws. Google claims Carbon has also secured the top position on the Artificial Analysis Intelligence Index v5.0, with a score of 59. It builds on Gemini 4 Argon, which Google introduced about two weeks ago. The company says the new model can handle complex engineering tasks that may take days to complete. However, Google is taking a cautious approach to the AI model’s rollout.
Google describes Gemini 4 Carbon as its “strongest model with state-of-the-art cybersecurity and long horizon capabilities” and says it delivers a “significant leap in cybersecurity capabilities.”
The tech giant said Gemini 4 Carbon brings improvements across the cybersecurity process, from finding weaknesses in software to developing fixes. It scored 88.2 per cent on SEC-bench Pro, a benchmark that tests how well AI models identify software vulnerabilities. It also achieved 54.8 per cent on ExploitGym, which measures the ability to test whether security flaws can be exploited.
Also read: Amazon cuts jobs again, sends early morning emails and texts to affected employees
I did say we’d be releasing our models periodically…introducing Gemini 4 Carbon, our next frontier model. Following Argon’s release just two weeks ago, Carbon shows the pace at which we’re pushing the frontier, it ranks #1 on the recently updated v5.0 of @ArtificialAnlys Intelligence Index at 59. It also leads across key cybersecurity evaluations scoring 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching. We’re seeing significant improvements in long-horizon tasks even compared to Argon, and we’ve strengthened frontier safeguards. As we did with Argon, we’re going to roll it out gradually + responsibly. Some trusted cyber defenders will get access starting this week. We’ll be expanding it to more partners through our Fairwind Program, ahead of wider availability soon, so hang tight!
— Sundar Pichai (@sundarpichai) October 28, 2026
Google also claimed Carbon performs better at generating secure code. “The Chrome team has reported that Carbon performed more than 10X better than another lab’s large frontier model when generating high quality patches,” it added.
The company’s internal tests showed a nearly 80 per cent drop in high-severity misalignment alerts in live coding environments. Google also said Carbon resisted all evaluated prompt-injection attacks, recording a 0 per cent attack success rate in the tested benchmark.
Google is initially making Gemini 4 Carbon available to a select group of cybersecurity experts. Access will gradually expand to other trusted partners through its Fairwind Program. The company said it is working through the US government’s voluntary process for testing the AI model before release.