The Ministry of Commerce is leading discussions to restrict overseas access to domestic large-scale AI models, and has also included technology leaks under the National Security Law. This signal is clear enough—core AI capabilities are to be locked within the wall, and the narrative of going global has completely shifted.

According to a Reuters report citing informed sources, Chinese authorities have held meetings over the past month with companies including Alibaba, ByteDance, and AI startup Z ai to discuss whether to restrict overseas access to China's most advanced artificial intelligence (AI) models, including those not yet released. The report said the meetings were led by China's Ministry of Commerce and covered discussions on closed-source models as well as more open model versions; officials also discussed the possibility of including the leakage or theft of proprietary AI technology under national security laws, and mentioned potential restrictions on sources of funding that can invest in domestic AI startups.