The Trump administration has reportedly asked OpenAI to stagger the release of its next major model, GPT-5.6, citing security concerns. OpenAI CEO Sam Altman confirmed this change during a company Q&A session on Wednesday, stating that the artificial intelligence firm would launch the system in a limited preview form first. Access during this initial phase would be restricted to a small group of enterprise customers only. The federal government would approve individual customers for access on a case-by-case basis throughout this preview period. This arrangement follows a similar request made to rival company Anthropic earlier this year.
While the delay affects the timeline for public availability, it ensures the model meets specific government security protocols before wider distribution. The approach allows OpenAI to maintain a commercial relationship with the administration while adhering to regulatory requirements. The company can still generate revenue from enterprise clients during the restricted window.
* GPT-5.6 enters limited preview before full public launch
* Government approval is required for each enterprise customer
* Anthropic faced stricter restrictions under the same policy
