The Trump administration has reportedly asked OpenAI to restrict the deployment of its forthcoming GPT-5.6 model, according to a report from Decrypt. The request follows a pattern of increasing government scrutiny over advanced AI systems, coming on the heels of similar limitations placed on Anthropic's Fable 5 and Mythos 5 models.
The specific nature of the requested restrictions remains unclear. Neither the White House nor OpenAI has issued an official statement confirming the report. Decrypt's sources indicate the administration's concerns may center on potential risks posed by the model's capabilities, though concrete details about GPT-5.6's features or training have not been disclosed.
This development aligns with broader regulatory efforts to assess and control frontier AI systems. The administration has previously signaled interest in evaluating large-scale AI models before they are made publicly available, citing national security and safety considerations. The reported request to OpenAI echoes actions taken against Anthropic, suggesting coordinated policy direction.
The AI industry now faces heightened regulatory uncertainty regarding model releases. OpenAI has previously engaged with policymakers on AI safety, but the reported request marks a more direct intervention by the executive branch. The company may face delays or modifications to its deployment timeline if compliance is required.
Competing AI labs are likely to face similar constraints. The administration's approach could set a precedent for pre-release review mechanisms across the sector, potentially reshaping how frontier models are introduced to the market. Critics argue such interventions may stifle innovation without clear safety benefits.