OpenAI launches Daybreak: GPT-5.5 cybersecurity platform to find software vulnerabilities

OpenAI has launched Daybreak, a new cybersecurity platform powered by GPT-5.5 and the Codex agent framework. Daybreak is designed to help tech companies identify and fix software vulnerabilities, and is a direct response to Anthropic's Project Glasswing.

The platform aims to embed cyber defense into software from the beginning of the development process, not just as an afterthought. Daybreak builds on the success of OpenAI's earlier GPT-5.4-Cyber model, which helped remediate over 3,000 security flaws.

It uses Codex Security to create an editable threat model from a company's software repository and focuses on the most realistic attack paths. It also validates fixes in an isolated environment, ensuring patches work without causing unintended issues.

The initiative positions OpenAI as a direct competitor to Anthropic's Project Glasswing, which major tech firms like Apple, Google, Microsoft, and Amazon have already adopted. OpenAI CEO Sam Altman said the company wants to work with as many organizations as possible to help them continuously secure their software. Three GPT-5.5 variants are available for different use cases, with stronger verification and access controls for specialized workflows.

As cybersecurity threats grow more sophisticated, Daybreak's proactive approach is seeking to redefine how software is developed and maintained. OpenAI is already working with industry and government partners to prepare for the next wave of cyber-capable models, signaling a long-term commitment to AI-driven security solutions.

With AI's continuous development and seemingly endless sophistication, security capabilities must keep pace with AI's intelligence. Anthropic and OpenAI are both attempting to calm that fear with their respective releases.