OpenAI releases GPT-6 Cybersecurity model
The story so far
OpenAI has released the cybersecurity model GPT-6 Cyber, claiming it can identify exploitable vulnerabilities, build usable exploit tools, and provide solutions for fixes, with training requirements to keep its behavior within authorized limits. The model was first opened to a small number of approved Daybreak Red customers through the API and Codex, and will be gradually expanded to all customers in the program; at the same time, GPT-5.6 Cyber is being opened to all Daybreak Red customers. According to OpenAI's system card, GPT-6 Cyber had a boundary violation rate of 0.07% in the ExploitGym honeypot evaluation without deploying security instructions, GPT-5.6 Sol was 49.5%, and it was 0% after deploying security instructions. The Decoder reported that 99.93% of tests did not cross boundaries without special security instructions, the previous generation was 94.49%; ExploitGym score was 48.5%, the previous generation GPT-5.6 Cyber was 32.1%. Supporting measures include abuse monitoring, credential protection, sandboxing, human supervision, and interception of high-risk behaviors targeting critical infrastructure. GPT-5.6 Cyber has previously been made available to approved customers through Amazon Bedrock in AWS GovCloud (US-East). OpenAI also claims to be working with Hacktron to find system vulnerabilities and strengthen AI defense.
AI-written from the reports · Updated 12 days ago
Developments
5 developments
- 6 Nov, 16:00 · 1 reportsDiscover vulnerabilities with HacktronOpenAI News:Partnering with Hacktron to find vulnerabilities in OpenAI’s systems
- 6 Nov, 03:06 · 1 reportsOpenAI released a new cybersecurity modelTechCrunch · AI:OpenAI released a new cybersecurity model. Is that enough?
- 5 Nov, 17:00 · 4 reportsGPT-6 CyberOpenAI News:GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents
- 5 Nov, 08:00 · 1 reportsGPT-6 Cyber System Card releasedOpenAI Deployment Safety:GPT-6 Cyber System Card
- 4 Nov, 15:00 · 1 reportsDaybreak Red available in AWS GovCloud (US-East)OpenAI News:Daybreak Red is now available in AWS GovCloud (US-East)
Coverage timeline
Follow the coverage to see every side of the story.
- OpenAI NewsPartnering with Hacktron to find vulnerabilities in OpenAI’s systems
OpenAI collaborates with Hacktron to find system vulnerabilities, test connections with third-party services, and leverage cutting-edge AI to enhance defenses.
- 36Kr · AIOpenAI has launched the cybersecurity model GPT-6 Cyber
OpenAI has launched GPT-6 Cyber, a cybersecurity model that discovers and verifies exploitable vulnerabilities. The first batch is being opened to Daybreak Red customers through API and Codex, with access gradually expanding.
- ITHome · AIFeaturedOpenAI releases cybersecurity-focused model GPT-6 Cyber, ExploitGym benchmark improvement of 51%
OpenAI releases cybersecurity-focused model GPT-6 Cyber, scores 48.5% in ExploitGym tests, a relative improvement of about 51% compared to GPT-5.6 Cyber's 32.1%
- The DecoderFeaturedOpenAIs GPT-6 Cyber can both break into systems and propose how to fix them
OpenAI launches GPT-6 Cyber, trained to find software security vulnerabilities, capable of discovering defects, building usable exploits, and suggesting patches. OpenAI claims it did not exceed boundaries in 99.93% of tests without special security instructions, compared to 94.49% previously; ExploitGym score 48.5%, previous GPT-5.6 Cyber was 32.1%.
- TechCrunch · AIOpenAI released a new cybersecurity model. Is that enough?
OpenAI discovered a severe vulnerability in Ubuntu servers through GPT 5.6 Sol, prompting the company to release the ChatGPT Cyber model for users of the Daybreak Red program, which was announced on November 6, 2026. The model had previously scanned the Linux kernel and found 948 potential issues. OpenAI has also invested $1 billion to support cybersecurity professionals in critical infrastructure, but the industry is concerned about the insufficient allocation of resources for cutting-edge models.
- OpenAI NewsFeaturedGPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents
OpenAI releases GPT-6 Cyber, a security model for discovering exploitable vulnerabilities and developing solutions, trained to keep its behavior within authorized limits.
- OpenAI Deployment SafetyFeaturedGPT-6 Cyber System Card
OpenAI releases GPT-6 Cyber system card, stating that the model's post-training focuses on cybersecurity capabilities and alignment, trained using reinforcement learning to replicate crashes, develop usable vulnerability exploits, and retrieve codebase vulnerabilities. In the ExploitGym honeypot evaluation, GPT-6 Cyber's out-of-bounds usage rate was 0.07% without deploying security instructions, GPT-5.6 Sol was 49.5%, and it was 0% with deployed security instructions.
- OpenAI NewsDaybreak Red is now available in AWS GovCloud (US-East)
Daybreak Red provides GPT-5.6 Cyber to authorized customers for cybersecurity work in AWS GovCloud (US-East) through Amazon Bedrock.