OpenAI releases GPT-6 Cyber security research model
OpenAI has launched GPT-6 Cyber, a model designed for security research that improves alignment while delivering stronger capabilities. The model supports previously announced Linux kernel research and is currently used by Mozilla, Palo Alto Networks, Calif, Cisco, and OpenAI's own security teams. Access is now available to select Daybreak Red customers and will be gradually expanded.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open AIHOT original