OpenAI introduced two Daybreak security products on November 12: Red Team for web application penetration testing and Defense Factory for investigating vulnerabilities and preparing patches.
Daybreak Red Team, available through early access, uses coordinated agents to examine application source code and context. The agents investigate attack paths and test whether reported vulnerabilities are reproducible.
The product accepts findings from code scanners, bug bounty submissions, and disclosure reports. After a fix, agents test whether the patch prevents the exploit. OpenAI says isolated environments and Guardian review restrict the applications and endpoints within an assessment’s scope.
Defense Factory handles the remediation work. Agents trace a vulnerability through a repository and develop a proposed patch. Engineers review and test the changes before merging.
Teams feed the service findings from existing security tools or Daybreak Red Team. Engineers also work directly with Codex to add context, inspect evidence, and refine a fix while background agents continue their investigations.
The tools cover different parts of the same workflow: Red Team checks vulnerabilities and patch effectiveness, while Defense Factory prepares changes for review. OpenAI also supports recurring checks and scans triggered by new commits.
Red Team’s product page invites applications for early access. Defense Factory directs prospective customers to sales. Neither page lists public pricing.
Comments (0)