OpenAI expands Daybreak with GPT-5.5-Cyber for automated patching

OpenAI introduced GPT-5.6-Cyber on Aug. 10 and said threat actors will increasingly use AI to conduct cyberattacks at unprecedented speed and scale, including in fully autonomous ways. The company paired the model with a broader Daybreak expansion aimed at end-to-end patch automation, framing the release as a response to a defense window that is narrowing as offensive AI gets more capable. The latest move builds on a cyber program OpenAI has spent months widening around trusted access, security workflows and outside partners.

In April, OpenAI said it was scaling its Trusted Access for Cyber program to thousands of verified individual defenders and hundreds of teams responsible for defending critical software. On April 23, OpenAI released GPT-5.5 and said its system card documented cybersecurity-related safety evaluations and targeted red-teaming for advanced risks. By May 7, OpenAI said Trusted Access for Cyber was being scaled with GPT-5.5 and GPT-5.5-Cyber, drawing a sharper line between general-purpose capability and tools designed for authorized security work.

OpenAI pushed that strategy further on June 22, saying Daybreak was being expanded with new tools, partnerships and the full version of GPT-5.5-Cyber to move from vulnerability discovery toward end-to-end patch automation. OpenAI said its models had already been used to discover and generate patches for critical vulnerabilities in major browsers, network infrastructure and operating systems including FreeBSD and the Linux kernel. OpenAI’s Daybreak pages describe the program as bringing together frontier cyber models, Codex Security, trusted workflows and ecosystem partnerships to help defenders find, validate and fix vulnerabilities before attackers can exploit them.

The enterprise side of that effort gained structure on June 14, when OpenAI launched the OpenAI Partner Network and said enterprise value depends not only on model capability but also on workflow redesign, systems integration and change management at scale. On June 22, Tenable said it was joining the OpenAI Daybreak Cyber Partner Program to explore how frontier AI could help organizations identify, prioritize and reduce cyber risk faster. IBM said the same day that it had joined the program to bring frontier AI into security operations and help enterprises counter machine-speed threats from Armonk, New York.

OpenAI said on Aug. 5 that it was expanding the Daybreak Cyber Partner Program to close the growing defense gap. Five days later, it introduced GPT-5.6-Cyber and new ways to unlock advanced cyber capabilities, underscoring how quickly Daybreak has moved from controlled access to a platform meant to automate more of the patching work defenders still struggle to keep up with.