The US government stated that advanced AI models such as GPT-5.6 will require government approval before release, shifting the system to a permit regime and expanding US AI export controls to include access rights to AI models. This exposes companies and public institutions relying on big tech models to policy-change risk. South Korea is building Korean AI models through the government’s Independent AI Foundation Model Project. OpenAI CEO Sam Altman disclosed talks with US officials, as Washington acted before the model’s release, following a pattern where the government previously intervened after Anthropic’s model release. The move is interpreted as a clear signal that Washington wants to control access to advanced AI models.