AI model releases · OpenAI

GPT-6 Cyber: OpenAI's cybersecurity model built to stay in scope

OpenAI has released GPT-6 Cyber, a model aimed at cybersecurity work. According to the company, it is built to find exploitable vulnerabilities, develop fixes, and stay within authorized scope while doing so.

The emphasis is on restraint. OpenAI describes the model as an aligned cybersecurity agent, and The Verge reports that the company says it is better at finding vulnerabilities and less likely to go rogue and hack other platforms along the way.

OpenAI says that in testing, GPT-6 Cyber went out of bounds less than 1 percent of the time. The Register phrases the same claim as avoiding exploitation of out-of-scope services over 99 percent of the time. Both figures are OpenAI's own.

GPT-6 Cyber is the latest version of OpenAI's cybersecurity model line and follows GPT-5.6 Cyber. It sits alongside the general GPT-6 and GPT-6.1 Terra among the lab's recent releases.

Key facts

  • Lab: OpenAI
  • Model: GPT-6 Cyber, a cybersecurity model
  • Purpose: find exploitable vulnerabilities, develop fixes, stay within authorized scope
  • OpenAI says it went out of bounds less than 1 percent of the time in testing
  • Previous cybersecurity model: GPT-5.6 Cyber

More model releases

Compiled by Newslog from the lab’s announcement and the coverage linked above. See every release in the AI model timeline, or follow today’s stories on the live AI page.