GPT-6 Cyber

OpenAI's security research model for finding, exploiting, and patching software vulnerabilities, limited to approved Daybreak Red security customers.

OpenAI researchers used it to build working exploits for 948 Linux kernel vulnerabilities and to find 81 previously unknown high-severity bugs in Firefox, which Mozilla has started fixing
Trained to stay inside the scope a researcher sets; in OpenAI's tests it avoided attacking out-of-scope services in 99.93% of runs without a safety prompt and 100% with one
OpenAI rates it Critical for cybersecurity risk under its Preparedness Framework, up from High for GPT-5.6 Cyber
Available through the API and Codex to a select group of approved Daybreak Red customers first, widening to all Daybreak Red customers soon; Daybreak Red API access requires IP allowlisting by default
GPT-5.6 Cyber is now open to all Daybreak Red customers, and Daybreak Blue alpha access for GPT-6 Astra is expanding to more organizations and individual researchers

Finding, proving, and patching vulnerabilities in large codebases with an AI agent.

Your security team runs authorized testing and needs a model that builds working exploits instead of refusing.

Security research and penetration testing teams, not marketing or general business users.

Join the newsletter to get ahead with AI.

I share strategies, new modules, free guides, and tools to get you better at using AI in your work.