China has rebuffed accusations from the United States that Chinese artificial intelligence developers are engaged in "aggressive, malicious" efforts to extract capabilities from leading American AI systems. Beijing has urged the U.S. to "refrain from making unfounded accusations or smears" and called for increased cooperation between the two AI powers.
The dispute comes ahead of planned talks between U.S. President Donald Trump and Chinese leader Xi Jinping later this month, where AI governance is expected to be a topic.
On Tuesday, the FBI, National Security Agency, and Cybersecurity and Infrastructure Security Agency issued a joint advisory alleging that Chinese AI developers had been extracting, or "distilling," capabilities from U.S. frontier AI models, including those from Anthropic, OpenAI, Google, and SpaceXAI, since at least late 2024. The advisory specifically named China's DeepSeek, Alibaba, Moonshot AI, and Z.ai, suggesting these activities were likely conducted with the awareness of the Chinese government.
According to the U.S. advisory, Chinese AI companies employ multi-path routing for distillation requests to gain unauthorized access, thereby violating the terms of service of U.S. AI companies. The agencies also stated that these companies achieve cost savings by bulk purchasing premium subscriptions from U.S. AI providers, sharing them among teams of developers.
In response, China's Ministry of Foreign Affairs spokesperson Mao Ning stated that China's AI development is a result of "high-level technological self-reliance and strength." She emphasized that both nations should strengthen cooperation to promote AI development that is "open, inclusive, universally beneficial and oriented toward the common good."
U.S. Treasury Secretary Scott Bessent reiterated the U.S. stance, stating that China can "never get ahead" of the U.S. in AI due to such distillation practices.
Chinese AI models, such as Moonshot's Kimi K3, have been gaining traction globally and are sometimes viewed as more cost-effective alternatives to U.S. models, even as U.S.-led restrictions limit China's access to advanced technologies. Previous accusations regarding AI capability extraction have also been made by Anthropic and OpenAI, claims that Beijing has consistently denied as groundless.