WASHINGTON, United States – The United States has warned that it could impose sanctions on Chinese artificial intelligence companies if investigations determine their open-source AI models were developed using stolen intellectual property from American firms.
U.S. Treasury Secretary Scott Bessent said the administration supports the development of open-source AI models but would not tolerate intellectual property theft. Okay News reports that Bessent said Washington would closely examine Chinese AI systems for evidence of unauthorized use of technology belonging to major U.S. developers before deciding whether sanctions are warranted.
The warning comes as Chinese AI companies continue to narrow the performance gap with leading American developers. Models such as Moonshot AI’s Kimi K3 have gained attention for their capabilities, increasing competitive pressure on firms including OpenAI and Anthropic. Reports have also suggested the Trump administration is considering broader restrictions on Chinese open-source AI models, although no final decision has been announced.
The debate has also revived questions over model distillation, a technique that transfers the capabilities of larger AI systems into smaller ones. While several U.S. AI companies argue that unauthorized distillation amounts to intellectual property theft, Microsoft CEO Satya Nadella recently questioned that view, saying it was inconsistent to support broad access to public data for AI training while opposing similar practices in model development.
The issue comes as AI companies worldwide continue to face legal scrutiny over how they train their models. Anthropic recently reached a $1.5 billion settlement over claims involving copyrighted books, while industry figures such as Hugging Face CEO Clem Delangue argue China’s progress is driven more by strong research capabilities and collaborative development than by copying American AI technology.
