A U.S. government website used an artificial intelligence model developed by Chinese tech giant Alibaba, raising fresh questions about Washington’s approach to Chinese AI technology.
Reuters reported that the Federal Register website, which is operated by the National Archives, offered an AI-powered search feature using Alibaba’s Qwen model.
The tool allowed users to browse public comments related to proposed federal regulations.
Its use drew attention because the FBI recently accused Alibaba of improperly copying technology from U.S. AI company Anthropic while developing its own artificial intelligence tools.
The Qwen-powered search option was removed Wednesday around the same time that social media posts began highlighting its presence, according to screenshots and archived source code reviewed by Reuters.
The National Archives and White House did not respond to Reuters requests for comment.
However, AI and cybersecurity experts cautioned that the presence of Qwen does not automatically mean sensitive U.S. government information was exposed.
The Federal Register primarily handles publicly available information, and Qwen is an open-weight AI model that can potentially be operated on an organization’s own infrastructure.
Senator Mark Warner said the level of security risk would depend partly on whether U.S. data remained inside government-controlled systems or was processed through infrastructure controlled by Alibaba.
The incident nevertheless highlights an unusual contradiction as Washington increasingly warns about potential security and technology risks associated with Chinese AI while a federal government website was itself using a Chinese-developed model.
The controversy comes amid intensifying competition between the United States and China over artificial intelligence, advanced chips and technological leadership.