Science & Tech · AI · 1 hr ago
OpenAI launches AI tools to find and fix software vulnerabilities
OpenAI has launched two AI-based cybersecurity products called Daybreak Red Team and Daybreak Defense Factory.
The tools were announced at a company event on November 12 and are designed for organizations that want to find and address weaknesses in their software.
Red Team tests web applications for vulnerabilities, using information about the application and its code to investigate possible ways in.
It can also check findings from security scanners, bug bounty programs and vulnerability reports.
Defense Factory investigates security findings and drafts patches, including for issues identified by other security products.
Engineers review and test the proposed fixes, and can create draft pull requests for GitHub; the patches are not applied automatically.
Red Team is accepting early-access requests, while Defense Factory is available as a preview plugin to ChatGPT workspaces with Daybreak access.
OpenAI has launched two AI-powered cybersecurity products designed to find vulnerabilities and help organizations fix them.
Daybreak Red Team uses AI agents to test web applications and investigate potential security weaknesses.
Daybreak Defense Factory investigates security findings and prepares proposed patches for engineers to review and test.
The products were announced at OpenAI’s Cyber at the Frontier event on November 12.
OpenAI is accepting requests for early access to Red Team, while Defense Factory is available as a preview plugin to workspaces with Daybreak access.
- Who
- OpenAI launched the products. The tools are intended for organizations and their security teams.
- What
- Daybreak Red Team tests web applications for vulnerabilities, while Daybreak Defense Factory investigates findings and prepares patches.
- When
- They were announced on November 12, 2026.
- Where
- At OpenAI’s Cyber at the Frontier event. The event’s location is not stated.
- Why
- To help security teams identify, validate, and address software vulnerabilities.
This story does not have two clearly opposing sides.
OpenAI announced Daybreak Red Team and Daybreak Defense Factory at its Cyber at the Frontier event.
OpenAI began accepting requests for early access to Daybreak Red Team and offered Defense Factory as a preview plugin to workspaces with Daybreak access.
- Products
- Daybreak Red Team and Daybreak Defense Factory
- Announcement date
- November 12, 2026
- Red Team function
- Tests web applications and investigates potential vulnerabilities
- Defense Factory function
- Investigates security findings and prepares proposed patches
- Patch approval
- Engineers review and test proposed patches before merge
3 sources
What people are saying
superhuman pen-test at the scale of automated continuous scanning
We're now at a point where companies will be secure only to the extent of the frontier model that they use to defend themselves and how well they utilize that
Agents use source code and app context to investigate attack paths and confirm vulnerabilities
Agents trace vulnerabilities through your code and develop patches. Engineers review and test proposed patches before merge