Photo courtesy of Anadolu
China rejected US allegations that Chinese artificial intelligence companies systematically tapped American AI models to acquire proprietary capabilities, calling on Washington to stop making what it described as false accusations.
Foreign Ministry spokesperson Mao Ning said Wednesday that China’s progress in artificial intelligence stemmed from domestic technological development and international cooperation, rather than the practices alleged by US authorities.
Mao said Beijing favored an open and inclusive approach to AI and argued that China and the United States, as major players in the technology, should pursue greater cooperation.
Her remarks followed accusations by a US cyber defense agency that several China-based AI developers had used large-scale knowledge distillation involving leading American AI systems since at least late 2024.
The US agency named DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun and Z.AI, alleging that the companies extracted billions of tokens through millions of interactions with frontier models including OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini and xAI’s Grok.
Knowledge distillation is a legitimate and widely used machine-learning technique in which a smaller model learns from the outputs of a more capable model. The dispute centers on allegations that proprietary systems were accessed or exploited at scale in ways their developers did not authorize.
US authorities alleged that the activity formed a significant part of the Chinese companies’ AI development strategies and was likely conducted with awareness of the Chinese government. The allegations have not been independently established, and Beijing rejected the characterization.
Mao urged Washington to follow through on understandings reached between the countries’ presidents and avoid actions that could further strain relations, while maintaining that greater US-China cooperation on artificial intelligence would benefit both sides.