U.S. Treasury Secretary Scott Bessent indicated that the Trump administration intends to investigate whether advanced artificial intelligence models originating from China have been derived or “distilled” using American intellectual property (IP). Bessent emphasized that the government does not support any form of IP theft.
Growing Concerns Over Global AI Competition
The increasing strength of open-weight Chinese models has raised alarms among both tech executives and government officials regarding the sustainability of U.S. leadership in the global artificial intelligence race. These domestic efforts are gaining ground against market leaders from American companies such as OpenAI and Anthropic.
For example, earlier this month, Moonshot AI, a Chinese startup, released Kimi K3, a model that reportedly surpasses competing firms across several industry benchmarks. Bessent explained during an appearance on “Mornings with Maria” that the technical process of illicit acquisition is known as distillation—a method where a smaller, less sophisticated model is constructed using the outputs generated by a stronger existing model.
“If we see, especially that overseas models are stealing from our great companies, we have the ability to sanction them because of this theft,” Bessent told Fox Business.
Investigations and Diplomatic Channels
Secretary Bessent stated that “we are finding watermarks of our U.S. large language models on many of the Chinese models, and that’s unacceptable.” He added that these findings would be subject to review in the coming days or weeks.
The concern regarding intellectual property theft has been formally raised previously; Anthropic sent a letter last month to the U.S. Senate Committee on Banking, Housing, and Urban Affairs alleging that Chinese technology firm Alibaba had performed what it described as “the largest known distillation attack” against its systems to date.
Furthermore, according to a report from Reuters published Tuesday, representatives from the U.S. and China are scheduled to hold talks concerning AI in September. Bessent is slated to represent the American side during these discussions.
History of IP Disputes in Tech
The issue of intellectual property infringement has impacted major players on both sides. Anthropic previously settled a class action lawsuit in September, agreeing to pay $1.5 billion after being accused by authors of illegally downloading books from databases containing pirated material. Separately, The New York Times filed a lawsuit against OpenAI and Microsoft in 2023, claiming that the companies infringed upon copyrights by utilizing their proprietary intellectual property for model training.
