On November 12 at its Cyber at the Frontier event, OpenAI unveiled two products that connect vulnerability discovery and remediation into a single agent pipeline.

Daybreak Red Team

In early access. An agent “pentester” for web applications: it uses the app's source code and context to validate attack paths, confirm findings from scanners and bug-bounty platforms, and then retest the patches. Not a one-off check but a closed loop: find, confirm, fix, verify.

Daybreak Defense Factory

Built on OpenAI's cyber models including GPT-6 Cyber. It scans connected GitHub repos, finds, validates and fixes vulnerabilities, opening a merge-ready pull request with a test. Available as a Codex plugin in research preview.

Why it matters

OpenAI president Greg Brockman put it plainly: “Companies will be secure only to the extent of the frontier model that they use to defend themselves.” Add the $1B Daybreak for Frontline Defenders subsidy. The signal for developers: security is becoming an agentic discipline, and manual audit is gradually moving into automated pipelines.