OpenAI has launched its Daybreak cybersecurity initiative, designed to help companies detect, analyze, and fix software vulnerabilities with the help of artificial intelligence.
According to an Engadget release on Monday (May 11), Daybreak was developed based on the premise that cyber defenses should be built into software from the start and not just focus on finding and fixing vulnerabilities.
Daybreak, according to the company, was introduced to prioritize addressing high-impact issues and reduce hours of analysis time to minutes to generate and test patches in the repository and send back results with audit-ready evidence to client systems.
OpenAI says Daybreak uses the company's proprietary artificial intelligence models, including a dedicated Codex Security agent that can be used to scan the codebase, validate the highest-risk findings, and fix them.
Daybreak will rely on the general-purpose GPT-5.5 AI model and GPT-5.5 with Trusted Access for Cyber for most defensive security workflows such as secure code review, malware analysis , patch validation, and threat detection engineering.
Meanwhile, the GPT-5.5-Cyber model is prepared for more specific needs such as cyber-attack simulation, penetration testing, and controlled validation of security systems.
