OpenAI will delay GPT-5.6 after Trump administration request
TL;DR
The Trump administration reportedly asked OpenAI to stagger GPT-5.6 instead of launching it broadly, citing possible security concerns around the new model. Sam Altman reportedly told employees in a company Q&A that GPT-5.6 will first ship as a limited preview for a small group of enterprise customers. During that preview, the US government would reportedly approve customer access case by case, turning the rollout into a politically supervised launch.
Nauti's Take
The key issue is not that a powerful model gets a slower rollout. The key issue is who controls the switch.
If the government approves customer access case by case, AI infrastructure starts to look like strategic infrastructure. That may reduce security risks, but it also creates a new power center between the state, OpenAI, and enterprise customers.
Briefingshow
This is bigger than a routine safety delay: the US government is reportedly inserting itself into the commercial rollout of a frontier model. For companies, access to the strongest AI systems may increasingly depend not only on readiness, price, or capacity, but on geopolitical classification and government approval.
Sources