China’s military researchers used outputs from advanced artificial intelligence models developed by OpenAI and Anthropic to train domestic AI systems designed to strengthen the country’s defense capabilities, according to a Reuters analysis of more than 80 Chinese academic papers and patents published on July 31, 2026.<\/p>\n
The investigation found widespread use of a technique known as model distillation<\/strong>, which allows developers to train smaller and more specialized AI models using the responses generated by larger systems. These compact models can then operate locally without the massive computing infrastructure required to build frontier AI models from scratch.<\/p>\n Reuters, citing research compiled by the Washington-based Jamestown Foundation, reported that the approach has been widely adopted by researchers linked to the People’s Liberation Army and other Chinese military institutions.<\/p>\n A 2024 paper from the National University of Defense Technology described the use of model distillation to shrink an image-processing model for deployment on unmanned aerial vehicles. The system enables drones to analyze video feeds in real time while supporting navigation and targeting decisions even when communications are disrupted.<\/p>\n Another paper authored by researchers from PLA Unit 96941, a military intelligence and cyberwarfare unit based in Beijing, detailed the use of OpenAI’s GPT-3.5 to process sensitive military source code. The resulting summaries were then used to train a domestic AI model capable of operating entirely within Chinese military networks.<\/p>\n Sunny Cheung, an associate fellow at the Jamestown Foundation who reviewed more than 60 of the papers, said Chinese military researchers are systematically attempting to capture the reasoning capabilities of Western AI models and adapt them for surveillance, cyberwarfare, and tactical decision-making.<\/p>\n He said the publications indicate efforts to transfer advanced reasoning from proprietary Western models into smaller systems that Chinese institutions can fully control and deploy domestically.<\/p>\n The findings add a new dimension to growing technology tensions between the United States and China ahead of discussions on AI governance and security. U.S. officials have accused some Chinese entities of using model distillation to extract capabilities from American AI systems, potentially bypassing export controls and undermining intellectual property protections. Chinese officials have rejected those claims, arguing that Washington is pursuing technological dominance in artificial intelligence.<\/p>\n Anthropic said it does not provide commercial access to its Claude models in China and warned that distilled models may lose the safety protections built into the original systems, potentially allowing sensitive capabilities to be transferred into models outside the company’s control.<\/p>\n Experts also cautioned that model distillation has technical limitations. Trevor Koverko, co-founder of AI data company Sapien, said distilled models remain less capable than the larger systems that train them. Rather than replacing frontier AI, he said, the technique transfers selected capabilities into lower-cost models that can be deployed and controlled locally.<\/p>\n Reuters reported that the White House, the Pentagon, China’s Ministry of Foreign Affairs, the People’s Liberation Army, and OpenAI did not respond to requests for comment.<\/p>\n","protected":false},"excerpt":{"rendered":" China’s military researchers used outputs from advanced artificial intelligence models developed by OpenAI and Anthropic…<\/p>\n","protected":false},"author":1,"featured_media":4846,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[16],"tags":[700,4594,4593],"class_list":["post-4848","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology","tag-artificial-intelligence","tag-military-technology","tag-model-distillation"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/posts\/4848","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/comments?post=4848"}],"version-history":[{"count":1,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/posts\/4848\/revisions"}],"predecessor-version":[{"id":4849,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/posts\/4848\/revisions\/4849"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/media\/4846"}],"wp:attachment":[{"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/media?parent=4848"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/categories?post=4848"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/zawaya24.com\/en\/wp-json\/wp\/v2\/tags?post=4848"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}