OpenAI has agreed to delay the release of its next-generation model, GPT-5.6, following a request from the Trump administration citing cybersecurity concerns, according to a report from The Verge.
The administration asked OpenAI to stagger the model’s rollout, and the government will now approve customer access on a case-by-case basis. The request aligns with a recent executive order advocating a 30-day pre-deployment review window for advanced AI models.
GPT-5.6 had not been officially announced by OpenAI, though prediction markets had anticipated a release by late June. Following the administration’s intervention, Polymarket odds shifted sharply: the probability of a June 30 release collapsed to roughly 8 percent, while markets now price a July release at over 80 percent.
The delay comes amid an increasingly tense relationship between the White House and major AI labs. The Trump administration previously imposed export controls on Anthropic’s frontier models, shutting down global access to Fable 5 and Mythos 5. Where that crackdown targeted China-bound technology transfer, the GPT-5.6 request is framed around pre-deployment safety review on home soil.
Observers are watching for OpenAI’s formal response and any revised timeline from CEO Sam Altman. The delay also raises unanswered questions about what specific security concerns prompted the request, the administration has not publicly detailed them.
Sources: OpenAI will delay GPT-5.6 after Trump administration request (The Verge, June 25, 2026); Trump administration requests OpenAI delay GPT-5.6 release (Crypto Briefing, June 25, 2026)