OpenAI is set to debut GPT-5.5-Cyber, a cutting-edge model designed to bolster cybersecurity, but it won't be available to everyone. CEO Sam Altman announced that the model will initially be released exclusively to a curated group of "trusted" cyber defenders, aiming to help critical institutions strengthen their defenses. The limited rollout is expected within days, following a similar strategy used by Anthropic with its Mythos model.
OpenAI Restricts Powerful Cyber AI to Elite Defenders
AI
April 30, 2026 · 1:00 PM