U.S. Lawmakers Introduce Bipartisan Bill to Ban Chinese AI Models in Government Agencies
A bipartisan group of U.S. lawmakers has introduced the “No Adversarial AI Act”, a bill aiming to prohibit U.S. executive agencies from using artificial intelligence models developed in China and other adversarial nations. The bill was introduced following a Reuters investigation that revealed DeepSeek, a Chinese AI firm, may be supporting China’s military and intelligence services while having access to significant quantities of Nvidia chips, raising national security concerns.
DeepSeek made headlines earlier in 2025 after claiming to develop a powerful AI model comparable to OpenAI’s ChatGPT, but at a fraction of the cost. This development sparked concern among U.S. policymakers and tech leaders, leading several companies and government bodies to voluntarily ban DeepSeek tools over data security risks. The Trump administration has also considered banning DeepSeek from U.S. government devices outright.
The bill was spearheaded in the House of Representatives by Rep. John Moolenaar (R-Mich.) and Rep. Raja Krishnamoorthi (D-Ill.), both top members of the Select Committee on the Chinese Communist Party. In addition to Chinese models, the bill includes AI systems from Russia, Iran, and North Korea. It proposes that the Federal Acquisition Security Council create and regularly update a list of restricted AI models from these countries.
Under the legislation, U.S. government agencies would be barred from acquiring or using AI models from listed adversaries unless granted a formal exemption by Congress or the Office of Management and Budget. There is also a provision allowing entities to challenge a listing if they can prove the AI model is not under foreign adversarial control or influence.
The bipartisan effort includes co-sponsors from both chambers of Congress, with Reps. Ritchie Torres (D-N.Y.) and Darin LaHood (R-Ill.) joining in the House, and Sens. Rick Scott (R-Fla.) and Gary Peters (D-Mich.) leading the bill in the Senate. Supporters of the bill stress the urgency of establishing a permanent firewall against potentially hostile AI systems that could infiltrate and compromise sensitive U.S. networks.
Source: Reuters