OpenAI rolled out Daybreak Red Team and Daybreak Defense Factory on November 12, 2026, according to The Register. Jessica Lyons, the outlet's Cybersecurity Editor, wrote that "the company rolled out Daybreak Red Team and Daybreak Defense Factory, which are both part of OpenAI’s Daybreak security platform." The Register reported that OpenAI President Greg Brockman spoke during a Thursday, November 12, 2026, keynote address. OpenAI's Daybreak Red Team page describes the first product as "AI penetration testing for web applications". Daybreak Defense Factory is presented as "AI Vulnerability Remediation", putting "AI agents to work investigating vulnerabilities and preparing patches for review." Both products are listed on OpenAI's Cyber at the Frontier page, which invites readers to "Explore the keynote, announcements, and guidance for the next era of cyber defense."

#What is Daybreak Red Team?

OpenAI describes Daybreak Red Team as "AI penetration testing for web applications" carried out by a coordinated team of agents. The company says: "Put a coordinated team of agents to work testing your web applications, validating vulnerabilities, and checking if patches prevent the exploit."

OpenAI says: "Agents use source code and app context to investigate attack paths and confirm vulnerabilities." Teams can also "Test findings from code scanners, bug bounty submissions, and disclosure reports" and "Use evidence to prioritize what to fix." After a change, OpenAI says customers can "Test whether a patch stops the exploit" and "Repeat the assessment as your apps change."

On scope, OpenAI says: "Enforce assessment scope with isolated environments and Guardian review. Control which apps and endpoints agents can reach."

The Register reported that "Red Team, which is available in early access, uses OpenAI models to test companies’ web applications for vulnerabilities." Nancy Wang, who the outlet says leads cyber-applied AI, described it as “superhuman pen-test at the scale of automated continuous scanning”, according to The Register. The Register did not name a specific model for Red Team.

#What is Daybreak Defense Factory?

OpenAI describes Daybreak Defense Factory as "AI Vulnerability Remediation". It says the product puts "AI agents to work investigating vulnerabilities and preparing patches for review." On the workflow, OpenAI writes: "Agents trace vulnerabilities through your code and develop patches. Engineers review and test proposed patches before merge."

OpenAI says findings can come from existing tools or from Red Team: "Bring findings from your security tools and Daybreak Red Team into the Defense Factory. Agents investigate each issue and develop a fix." The Red Team page points to Defense Factory under the heading "Turn verified findings into fixes".

Users can work on a patch in Codex, OpenAI says: "Add context, examine evidence, and tune a patch with Codex. Test the changes while agents continue working in the background." For ongoing checks, OpenAI says: "Scan new commits or schedule recurring checks as your code changes. Put agents to work investigating new findings." To start, it tells teams to "Connect your repositories and security findings to start turning your backlog into fixes."

The Register reported that Defense Factory automates the “defensive loop” by “finding, validating, and fixing vulnerabilities at scale.” The outlet wrote that this product "also uses agents built on OpenAI’s latest cyber models" currently, "that’s GPT-6 Cyber." According to The Register, "Engineers review and approve the proposed fixes before they go into production." Wang said, according to the outlet, “This way engineers get a merge-ready pull request with the patch and a test that shows that it works”.

#Who can use it, and how do they get it?

The Register reported that Daybreak Red Team "is available in early access". OpenAI's Cyber at the Frontier page includes the prompt "Request Daybreak Access". The Defense Factory page tells teams to "Connect your repositories and security findings to start turning your backlog into fixes."

OpenAI's product pages do not state an availability tier, a price, or a general availability date for either product, and they do not say how Defense Factory is offered. The Register's early-access line applies to Red Team.

#How does it compare?

Daybreak Red Team is a new product. It is distinct from Daybreak Red, the earlier access program for OpenAI's cyber models. We covered that program in our reports on GPT-6 Cyber for Daybreak Red defenders and OpenAI's Hacktron partnership, which expanded Daybreak Red access. No source says the new Red Team product replaces that program.

At the keynote, OpenAI President Greg Brockman said, according to The Register, “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”. The Register also reported that he said security “fundamentals are important, and not to be discounted.” He added, according to the outlet, that “you need to really reorient your security practice around what these models can do”.

Our analysis: The launch packages agent testing and patch preparation as two named products on the Daybreak platform, with engineer review before merge in OpenAI's account and before production in The Register's. The Register's GPT-6 Cyber detail applies to Defense Factory only. OpenAI's capability descriptions are vendor claims; the sources include no independent test of either product.

#What is still unclear?

OpenAI's product pages do not carry a publication date, a price, or a named model. The Register is the source for the November 12, 2026 rollout date, for Red Team early access, and for naming GPT-6 Cyber on Defense Factory. How Defense Factory is offered, and whether it has a general availability date, is not stated in the sources. OpenAI's Red Team page mentions "Guardian review" without defining Guardian. Beyond "OpenAI models", The Register does not name the model used by Red Team.