Chinese military researchers have used outputs from leading US artificial intelligence models developed by OpenAI and Anthropic to train domestic AI systems advancing China’s defense capabilities, according to a Reuters review of more than 80 Chinese academic papers and patents.
The findings reveal widespread use of “model distillation,” a technique where outputs from powerful AI systems are used to train smaller, specialized models that can be deployed locally without the enormous computing resources needed to build frontier systems from scratch. The practice has raised alarms in Washington about unauthorized extraction of AI capabilities despite chip export controls.
A paper by researchers in PLA Unit 96941, a military intelligence and cyber-warfare unit in Beijing, described using OpenAI’s GPT-3.5 to process sensitive military source code. The researchers used GPT-3.5 to summarize software code and trained a domestic model on those summaries to run entirely within Chinese military networks.
At the North University of China, which has close links to the country’s weapons industry, researchers used Anthropic’s Claude 3 Haiku to generate synthetic training data for social media monitoring and content moderation. Anthropic said it does not provide commercial access to Claude in China and monitors for policy violations.
A 2024 paper from the PLA’s National University of Defense Technology described using distillation to shrink an image-processing model for deployment on unmanned aerial vehicles, allowing drones to analyze live video for navigation and targeting in real time. Researchers at China’s Academy of Military Sciences used distillation to run target-recognition models on tactical hardware during simulated maritime operations.
Sunny Cheung, a Jamestown Foundation fellow who analyzed over 60 of the papers, said Chinese military scientists are systematically capturing the reasoning steps of Western models. “Teaching a model the right answer is one thing, but teaching it the reasoning behind the answer is much harder,” Cheung said.
The issue has become a major flashpoint ahead of US-China talks on AI governance. US officials accuse Chinese entities of using distillation to extract capabilities from American models, potentially undermining export controls. China has rejected the accusations, calling Washington’s approach AI “hegemonism.”
Experts caution that distilled models have significant limitations and cannot fully replicate the broad intelligence of frontier systems. Trevor Koverko of AI data company Sapien said distilled models remain less capable than their teacher models. “It is best understood as transferring selected capabilities into a cheaper, locally controlled system, not achieving independence from frontier AI.”
The White House, Pentagon, China’s foreign ministry, the PLA, and OpenAI did not respond to requests for comment.
Sources:
– Reuters: Chinese military researchers tap US AI models to train defence systems
– Rappler: Chinese military researchers tap US AI models to train defense systems
discussion