OpenAI Introduces GPT-6 Cyber for Vulnerability Detection

OpenAI has introduced GPT-6 Cyber, a cybersecurity model trained to find and validate exploitable vulnerabilities while staying within the scope researchers set.

Benchmark Performance

On ExploitGym, which tests whether models can exploit a specified vulnerability, GPT-6 Cyber scores 48.5%, up from 32.1% for GPT-5.6 Cyber. The model also achieved 40.3% recall of vulnerabilities in a new evaluation for offensive web penetration testing, compared with 32.0% for GPT-6 Astra.

Alignment and Scope

In an alignment evaluation, GPT-6 Cyber avoided exploiting out-of-scope services in 99.93% of tests without its deployment prompt, and 100% with this prompt. This is an improvement over GPT-5.6 Cyber, which avoided exploiting out-of-scope services in 94.49% of cases without deployment safety prompts.

Real-World Impact

OpenAI researchers used GPT-6 Cyber to develop 948 privilege escalation exploits in Linux and identify 81 previously unknown high-severity vulnerabilities in Firefox. Mozilla has released fixes for many of the Firefox vulnerabilities. Security teams at Cisco, Calif, and CodeRabbit are using it for vulnerability research and code review.

Access and Safeguards

GPT-6 Cyber is initially available through the API and Codex to a select group of approved Daybreak Red customers. OpenAI has classified the model as Critical in cybersecurity under its Preparedness Framework, up from High for GPT-5.6 Cyber.

Why it matters

The release expands access to advanced cybersecurity research tools for organizations, independent researchers, and open-source maintainers through the Daybreak program, while introducing new safeguards for critical infrastructure and misuse detection.

---

Sources

• openai.com: GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents | OpenAI

This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.