Shafaqna English– A Reuters investigation, which examined more than 80 Chinese academic papers and patent filings, has revealed that military researchers in China have been utilizing outputs from prominent U.S. artificial intelligence models—specifically those developed by OpenAI and Anthropic—to train their own domestic AI systems. The objective of this effort is to enhance China’s national defense capabilities.
These previously undisclosed findings provide a rare and revealing window into the methods by which Chinese institutions linked to the military and national security sectors are exploiting advanced U.S. AI models as a convenient shortcut to develop their own specialized systems.
This practice continues despite Washington’s ongoing efforts to curb Beijing’s access to state-of-the-art semiconductors and other strategically sensitive technologies.