WASHINGTON [Sept. 9]: The United States has accused several Chinese artificial intelligence companies of systematically using technology developed by American AI firms to accelerate their own AI systems, adding a fresh point of tension ahead of a planned meeting between U.S. President Donald Trump and Chinese President Xi Jinping later this month.
U.S. law enforcement and intelligence officials said six Chinese companies, including DeepSeek, Moonshot AI and Alibaba, had used a technique known as AI model distillation to develop their products more rapidly and at lower cost.
Distillation involves training a smaller AI model using outputs generated by a larger and more expensive model. The approach can reduce the resources and time required to develop new systems, but U.S. officials allege that Chinese companies have used the technique to obtain and replicate capabilities from leading American AI models.
According to the officials, the Chinese companies targeted systems developed by Anthropic, OpenAI, Alphabet’s Google and SpaceX. They said the activity was carried out “likely with Chinese government awareness.”
“China-based AI companies are engaging in aggressive, malicious and targeted distillation activities at an industrial scale,” U.S. officials said.
The allegations come as Washington and Beijing prepare for a series of high-level engagements. The Trump administration is preparing for Xi’s expected visit to the United States in late September, while the two countries are also expected to hold discussions on artificial intelligence safety risks during a dialogue planned for mid-September.
China rejected the U.S. accusations.
Chinese Foreign Ministry spokesperson Mao Ning said Wednesday that the country’s progress in artificial intelligence reflected its efforts toward “achieving high-level science and technological self-reliance.”
“We hope the U.S. will earnestly implement the important consensus reached by the leaders of both countries and refrain from making false accusations and smearing China,” Mao said.
She also called for greater cooperation between the two countries, saying, “As major powers in artificial intelligence, China and the U.S. should strengthen cooperation.”
Washington has previously raised similar concerns about China’s development of AI technology. In April, U.S. officials made a comparable accusation ahead of Trump’s visit to Beijing.
The latest U.S. statement also linked the alleged use of AI model distillation to national security concerns. Officials said the practice could not only lower research and development costs for Chinese companies but also strengthen China’s “military and cyberattack capabilities that could be used against the U.S. and our allies.”
Chinese military researchers have also been reported to have used outputs from leading U.S. AI models to train domestic systems and support the development of defense capabilities.
