U.S. Treasury Secretary Scott Bessent announced Tuesday that the United States will increase scrutiny of Chinese open-source AI models [2].

This move signals a tightening of trade and technology restrictions between the two superpowers. By targeting open-source models, the U.S. aims to prevent the proliferation of artificial intelligence built on stolen corporate secrets, which could erode the competitive advantage of American tech firms.

Bessent said the Treasury Department will examine models created overseas, specifically focusing on those from China. The administration intends to take action if evidence emerges that these models utilize intellectual property stolen from U.S. companies [1].

"We will carefully examine open source AI models created overseas, including those from China, and take action on any evidence of intellectual property theft from US companies," Bessent said [1].

The Secretary said the U.S. is prepared to use financial penalties to enforce these standards. He said he is ready to implement restrictive measures against offending entities [3].

"I have the ability to sanction, and I'm not afraid to use it," Bessent said [3].

These threats come as the U.S. and China continue to navigate complex diplomatic waters regarding technology. While tensions remain high, follow-up AI talks are already scheduled for September 2026 [4].

Separate from the policy announcement, reports indicate that Zhongji Innolight plans to raise $7 billion in a Hong Kong listing [5]. This financial activity occurs amid the broader backdrop of shifting regulatory environments for Chinese technology companies.

“"I have the ability to sanction, and I'm not afraid to use it."”

The U.S. is expanding its definition of national security to include the provenance of open-source weights and architectures. By threatening sanctions over intellectual property theft, the Treasury is leveraging financial warfare to discourage Chinese labs from utilizing leaked or stolen U.S. data, potentially slowing the development of Chinese AI while creating a more rigorous compliance burden for global open-source collaborators.