At the request of the Trump administration, OpenAI is postponing the wide release of GPT-5.6, allowing only select enterprise customers to preview the technology under stringent government oversight.
- GPT-5.6 release limited to select enterprise users initially
- Trump administration to approve customer access case-by-case
- Regulatory actions differ for OpenAI and Anthropic
What happened
OpenAI’s next major AI model, GPT-5.6, will not be broadly available immediately following its development. Instead, OpenAI will offer access in a limited preview form to a small group of enterprise customers. This decision complies with a request from the Trump administration, which seeks to manage and oversee the deployment due to potential security risks associated with advanced AI systems.
During this controlled preview phase, the Trump administration will review and approve access for individual customers on a case-by-case basis. This regulatory process represents a cautious approach, aiming to balance innovation with national security concerns amid rapid AI advancements.
Why it matters
The Trump administration’s intervention signals a growing willingness by regulators to assert control over AI technologies that could pose security or geopolitical risks. The staggered release approach for GPT-5.6 marks a shift from earlier, more permissive attitudes within the US government that emphasized speed and competitiveness in AI development and export.
Additionally, the government’s distinct treatment of OpenAI, compared with its more restrictive directive to Anthropic—requiring suspension of foreign nationals’ access—demonstrates how regulatory strategies may vary across companies. This uneven landscape introduces new uncertainties for AI developers and users navigating compliance and operational risks.
What to watch next
Further, any responses or adaptations by competitors like Anthropic, given their stricter limitations, will be telling of how these regulatory interventions shape competition and collaboration within the AI sector. Broader implications for international AI exports and security frameworks are also expected to emerge from these unfolding regulatory dynamics.