China says US accusations of AI theft are ‘groundless’

The United States recently accused several Chinese AI labs of stealing American model capabilities through large-scale knowledge distillation campaigns. However, China vehemently denied these allegations, calling them "groundless" and accusing the US of engaging in double standards. According to China's Ministry of Commerce, the US approach aims to create industrial monopolies, and the accusations are a pretext to gain an unfair technological advantage.

Distillation, a technique where a smaller AI model learns to mimic a more powerful one, is widely used in the industry. However, the US claims that Chinese firms have scaled this practice to illicitly copy American AI systems. The US cyber defense agency's advisory noted that Chinese companies like DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI allegedly used millions of interactions to steal capabilities from US models, including those from OpenAI, Anthropic, Google, and xAI.

This strategy reportedly allowed Chinese companies to develop frontier models more quickly and at lower costs. The accusations gained attention ahead of a meeting between US President Donald Trump and Chinese President Xi Jinping, where AI could be a key discussion point. China's foreign ministry spokesperson, Mao Ning, responded by urging the US to refrain from making unfounded accusations against China.

She emphasized China's self-reliance and technological strength in the AI field and called for stronger cooperation between the two nations.