OpenAI introduces GPT-6 Cyber for scoped vulnerability research
Original title: GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents
OpenAI reports 48.5% on ExploitGym, up from 32.1%, and 99.93% in-scope behavior without the deployment prompt.
OpenAI introduced GPT-6 Cyber, a model trained to find and validate exploitable vulnerabilities inside limits researchers set. The company says it helped develop 948 Linux privilege-escalation exploits and find 81 previously unknown high-severity Firefox flaws, and that Mozilla has released fixes for many of them. OpenAI rates the model Critical under its Preparedness Framework and is offering deeper testing through Daybreak Red.
Unconfirmed: Benchmark scores, exploit counts, and the Critical rating are OpenAI's own figures.
- OpenAI says web-pentest recall was 40.3% for GPT-6 Cyber versus 32.0% for GPT-6 Astra.
- On RemoteExploitBench, GPT-6 Cyber completed 95% of tasks while Astra with Daybreak Blue refused all.
- With the deployment prompt, OpenAI reports 100% avoidance of out-of-scope exploitation.
- Cisco, Calif, and CodeRabbit are using the model for research and code review.
- Vendors
- OpenAI
- Organizations
- MozillaCiscoCodeRabbitCalif
- AI models
- GPT-6 CyberGPT-5.6 CyberGPT-6 Astra
From the original article
13 min read at openai.comIntroducing GPT‑6 Cyber, OpenAI’s cybersecurity model for finding exploitable vulnerabilities and developing fixes, trained to stay within authorized scope.
openai.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.