# OpenAI releases GPT-6 Cyber for approved Daybreak Red customers

> Original title: OpenAI released GPT-6 Cyber, a model built to find and exploit software vulnerabilities without straying outside authorized targets.

- Source: Rohan Paul
- Published: 2026-11-05T22:27:59.000Z
- Bad Influence: https://badinfluence.news/items/pe91k7cypcofnemmmlolrlof3
- Original: https://x.com/rohanpaul_ai/status/2118470407313646054

## Summary

OpenAI released GPT-6 Cyber, a security research model that is currently available only to select approved Daybreak Red customers through the API and Codex. OpenAI says it confirmed 1,724 Linux kernel bugs and built working attacks for 948 of them, versus 166 for GPT-5.6 Sol, and found 81 serious Firefox bugs. OpenAI rates the model Critical, its highest cyber-risk level, and reports it stayed within authorized scope in 99.93% of ExploitGym Honeypot tests without its deployment prompt.

## Post

OpenAI released GPT-6 Cyber, a model built to find and exploit software vulnerabilities without straying outside authorized targets.

Right now only a select group of approved Daybreak Red customers can use it, through the API and Codex.

> On real software, it confirmed 1,724 Linux kernel bugs and built working attacks for 948 of them, while GPT-5.6 Sol, tested the same way, managed 166.

> It also found 81 serious Firefox bugs nobody knew about, and Mozilla has already released fixes for many of them.

> Because the model is this good at attacking, OpenAI rates it Critical, its highest cyber-risk level

> That gain came with training to keep the agent inside the scope a researcher authorizes.
ExploitGym Honeypot tests that by planting easy-to-hack side services, such as a package registry, whose compromise also reveals the task’s flag.

Without its deployment prompt, the model ignored that bait in 99.93% of tests, against 94.49% for its predecessor, and in 100% with the prompt.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWZNLKIaYAAlTWt.jpg&mode=thumb&exp=1794182400&sig=06ec54c344635684)

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWZNNPybIAAxVom.jpg&mode=thumb&exp=1794182400&sig=a91752dbefd84d64)

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWZNQFzaUAA9WAo.jpg&mode=thumb&exp=1794182400&sig=2926b7dca78d9f71)

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWZNRdjbQAAKzhV.jpg&mode=thumb&exp=1794182400&sig=8f68b37f05561616)

## Quoted post @OpenAI

> Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.
> 
> The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.
> 
> GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.
> 
> https://openai.com/index/introducing-gpt-6-cyber/

https://x.com/OpenAI/status/2118390190246879554