OpenAI rolls out GPT-5.6-Cyber to bolster AI-powered cyber defense

OpenAI introduced GPT-5.6-Cyber and expanded Daybreak to give cybersecurity defenders broader access to its cyber-focused models, splitting Daybreak into Blue (no system-level guards) and Red (GPT-5.6-Cyber for exploit validation and advanced research). The program enables partners like Accenture, IBM, CrowdStrike, Cisco, and Palo Alto Networks to embed the models in security products and services. In testing, GPT-5.6-Cyber handled about 95% of advanced cybersecurity prompts, vs. much lower responses from GPT-5.6-Sol and Daybreak Blue, and the rollout follows recent scrutiny after a Hugging Face incident, signaling a push to expand defender-focused AI tooling.