# OpenAI limits US rollout of new GPT-5.6 model > Source: > Published: 2026-06-26 17:25:05+00:00 ### What happened **OpenAI** launched a limited, US-only preview of its newest model, GPT-5.6, and delayed a broader public rollout at the request of the U.S. government, Reuters reports. Reuters says the initial access is restricted to a small set of vetted partners and that OpenAI shared partner details with federal agencies. Per Reuters, OpenAI stated in a blog post: "We are taking this short-term step because we believe it is the strongest path to broader availability in the coming weeks, while we work with the Administration to develop the cyber Executive Order framework and a repeatable process for future model releases." ### Government involvement **Axios** reports the request came from the Office of the National Cyber Director (ONCD) and the Office of Science and Technology Policy (OSTP), citing national security concerns around the model's cybersecurity capabilities. **The Information** and **TechCrunch** report that CEO Sam Altman told employees the administration would approve customer access "customer by customer" during the preview period. **Reuters** reports government interest is tied specifically to cyber threats and potential misuse risks. ### Context and significance Multiple outlets including TechCrunch, The Verge, The Information, Reuters, Axios, and CNN frame the decision as part of an intensifying U.S. government engagement on frontier AI. This follows the Commerce Department's June 2026 export-control directive requiring Anthropic to suspend access to its Fable 5 and Mythos 5 models, per CNBC -- a pattern that suggests government pre-release review is becoming a recurring feature of frontier model launches rather than a one-off event. ### What to watch Three indicators to monitor: whether the U.S. government formalizes review procedures under the cyber Executive Order OpenAI referenced; how many and which partners receive early access and whether the list expands; and how quickly OpenAI transitions from the staged preview to general availability. **Reuters** notes OpenAI cautioned that government access and oversight should not become a permanent standard -- language that could shape future public debate and regulatory design around frontier model launches. ### Practical implications For developers and security teams, staged previews limit hands-on evaluation. Controlled rollouts concentrate early feedback in a smaller set of organizations, accelerating detection of misuse vectors but slowing broader community red teaming and integration testing. For policy watchers, the episode advances a precedent where the U.S. government can request pre-release access to frontier models and veto individual customer deployments, influencing how companies and regulators negotiate launch protocols going forward. ## Key Points - 1OpenAI limited `GPT-5.6` to a US-only, partner preview after the Office of the National Cyber Director and OSTP requested early access, Reuters and Axios report. - 2The administration will vet customer access case-by-case during the preview, according to reporting by The Information and TechCrunch, citing CEO Sam Altman. - 3Government pre-release review of frontier models appears to be shifting from an exception to a recurring pattern, following a similar directive applied to Anthropic in June 2026. ## Scoring Rationale Government-requested restriction on a frontier model release establishes a regulatory precedent that directly affects model availability, independent red teaming, and enterprise procurement timelines. The story combines a major model rollout with the most direct U.S. government intervention in AI deployment documented to date. Practice interview problems based on real data 1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with. [Try 250 free problems](/problems)