Google debuts flagship Gemini 4 Carbon model with cybersecurity upgrades

SHARE UPDATED 19:27 EDT / OCTOBER 28 2026

Google LLC today debuted Gemini 4 Carbon, a frontier large language model adept at complex programming tasks.

The LLM is rolling out just two weeks after the search introduced its previous flagship LLM, Gemini 4 Argon. According to Google, internal tests have shown Carbon to be more than twice as good at generating secure code. Additionally, the model outperforms GPT-6 Astra and Claude Opus 5.5 across multiple benchmarks.

Carbon’s most notable benchmark result is its record score on the Artificial Analysis Intelligence Index. It’s a test that combines more than a half-dozen popular LLM evaluations. Google tested Carbon on a new version of the index that was publicly released today. It includes two additional benchmarks and removes a number of saturated questions, questions deemed too easy for frontier models.

Carbon is particularly adept at cybersecurity tasks. It handily beat the previous record on SEC-bench Pro, which evaluates LLMs’ ability to find weak points in complex software systems. The model is also better than its predecessor at patching the vulnerabilities it finds.

“The Chrome team has reported that Carbon performed more than 10X better than another lab’s large frontier model when generating high quality patches,” Koray Kavukcuoglu, the senior vice president of Google DeepMind and Google’s chief artificial intelligence architect, wrote in a blog post.

Argon, the company’s previous flagship LLM, includes guardrails designed to mitigate hallucinations and cyberattacks. Google implemented an improved version of that feature set in Carbon. According to the company, a number of internal teams that use Carbon in “live coding environments” reported a nearly 80% drop in misalignment alerts.

When LLMs struggle to answer a user question, they sometimes fill their knowledge gaps by hallucinating inaccurate information. Google configured Carbon to avoid such behavior.

“In our internal testing, Carbon cuts unsupported claims by approximately 2x over Argon,” Kavukcuoglu wrote. “It puts methodological caveats directly into the deliverable or withholds blocked actions with a clear list of what remains required, rather than fabricating or hedging around missing evidence.”

Alignment issues can also crop up while a model is being trained. Developers train frontier LLMs by giving them a large number of tasks similar to the ones they’ll be expected to perform in production. LLMs sometimes cheat on those tasks, which is not conducive to the learning process.

Google tested whether Carbon would use shortcuts to solve the coding challenges in the Terminal-Bench 4.0 benchmark. The model attempted to cheat across only 0.4% of the questions and none of the attempts were successful. Carbon performed even better on the Gray Swan IPI Benchmark, which tests LLMs’ susceptibility to indirect prompt injections. The model blocked all of the simulated prompt injections.

Google will make Carbon available later this week through its Fairwind Program, which gives organizations such as internet providers early access to its LLMs. Participants use the models to fix cybersecurity flaws in their systems. Google plans to refine Carbon’s guardrails based on feedback from early adopters before making it broadly available.

Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.

Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.

F5 and Oracle plan to bring F5 web app and API protection to Oracle Cloud Infrastructure

Fortinet plunges 26% after-hours and takes Tenable with it as Gen Digital rises

Experian builds cyber resiliency into its AI infrastructure

Google to launch secure multi-party computing service for crypto wallets

AI cybersecurity shifts to governing agents with Okta identity

Why InfoScale wants enterprises to protect live operations, not just servers

SECURITY - BY DUNCAN RILEY. 26 MINUTES AGO

SECURITY - BY DUNCAN RILEY. 47 MINUTES AGO

BLOCKCHAIN - BY MARIA DEUTSCHER. 2 HOURS AGO

SECURITY - BY JONATHAN ANTHONY. 2 HOURS AGO

SECURITY - BY SLOANE KALI FAYE. 3 HOURS AGO

© 2026 SiliconANGLE Media Inc. All rights reserved.