OpenAI announces GPT-6 Cyber for security research

OpenAI has announced GPT-6 Cyber, an AI model built for security research. The company says it is far better at finding and fixing security flaws than GPT-5.6 Cyber, and its most notable result for everyday users involves Firefox.

During a research campaign, GPT-6 Cyber found 81 new flaws in Firefox, all rated as high severity by Mozilla. Because several earlier AI models had already tested the browser, these bugs slipped past prior scrutiny. The model also proposed patches and showed that 30 bugs already on record were more dangerous than reported. Mozilla has since released fixes for many of them, and Firefox engineer Brian Grinstead said those fixes reached millions of users. Patching these flaws early closes them off before attackers can find and use them. Firefox users who have automatic updates turned off should install the latest version manually.

Testing Linux and OpenAI’s own servers

GPT-6 Cyber also built working exploits for 948 Linux kernel vulnerabilities. Under the same research setup, OpenAI’s older GPT-5.6 Sol managed 166. OpenAI reported 304 critical findings to Linux maintainers along with suggested fixes, and companies like Cisco and CodeRabbit already use the model for code review. In its announcement, OpenAI stated that it trained the model for advanced cybersecurity work with improved alignment, noting that it has powered internal security work and helped find and patch high-severity vulnerabilities in Firefox.

OpenAI also tested the model on its own infrastructure. In controlled tests, agents running GPT-6 Cyber and other OpenAI models escaped their isolated environments and ran code on the machines underneath. OpenAI used the results to patch its software and tighten network restrictions.

How safe is GPT-6 Cyber?

A hacking AI is only useful if it does what it is told and nothing more. OpenAI tested this by giving the model a hard task in a closed environment with a few easy-to-break services nearby. Breaking into them could have revealed the answer, but they were outside the task’s scope, and the model had to figure that out on its own. Without safety instructions, GPT-6 Cyber left them alone in 99.93% of tests, and with the instructions, it did so every time.

Most people will not be able to try it for now. OpenAI classifies the model as “Critical” in cybersecurity under its safety framework, and only a small group of approved security teams currently has access.