The Trump administration has requested that OpenAI restrict the release of its upcoming model, GPT-5.6, to a select group of government-approved partners prior to a broader rollout, citing security concerns. This request represents the first instance in which the U.S. government has proactively asked an American AI company to limit a model's launch before its release.
The White House's Office of the National Cyber Director and the Office of Science and Technology Policy made the request as the administration develops a framework for testing and evaluating the security of new AI models. OpenAI CEO Sam Altman reportedly communicated the plans for a limited rollout in a memo to employees, stating that the company prefers a more sustainable approach for future releases.
Furthermore, OpenAI has been collaborating with the administration on the model release since before Anthropic revoked access to its models due to a Commerce Department directive. Altman discussed GPT-5.6 with Commerce Secretary Howard Lutnick, who emphasized the importance of government testing and approval of the model. The government intervened due to the model's advanced capabilities, ensuring that adequate safeguards are in place.
Earlier this month, President Trump signed an AI security executive order directing several agencies to establish a voluntary testing protocol for AI companies before releasing new models. Political disagreements over the program's restrictions delayed the executive order. AI labs are facing challenges as they strive to release new models while competing with increasingly advanced Chinese open-source models, amid growing concerns about the potential misuse of powerful AI technologies.