OpenAI · 2026-09-24 · rumor
GPT-6 Cyber rumor — OpenAI may preview its next security model at DevDay
Fortune reports that OpenAI will preview GPT-6 Cyber within days, together with a new product for deploying it safely. Daybreak Red customers are already alpha testing it. OpenAI has not commented.
Fortune says OpenAI's fourth security model of 2026 is weeks or days away, with a deployment product attached.
What is it?
GPT-6 Cyber is a cybersecurity model that OpenAI plans to preview soon, according to Fortune senior AI reporter Emily Forlini, who cites multiple unnamed sources familiar with the plans. The report also describes a yet-to-be-named product that helps customers deploy the model securely and automate workflows that patch vulnerabilities.
How does it work?
Per the report, a limited group of customers in OpenAI's application-only Daybreak Red program already has GPT-6 Cyber in alpha testing. The current public model in this line, GPT-5.6 Cyber, is available only to approved defenders through the Daybreak program and only on the Responses API, so a GPT-6 version would most likely follow the same gated access.
Why does it matter?
Security teams on Daybreak would get a GPT-6-class model for authorized vulnerability research, after GPT-5.4, 5.5 and 5.6 Cyber earlier in 2026. Confidence: medium. The sourcing is anonymous but comes from a named Fortune reporter, and Reuters picked it up. OpenAI did not respond to Reuters. A preview at DevDay on September 29 would confirm it; no announcement there would weaken the claim.
Who is it for?
security teams, Daybreak program members