Amid heightened geopolitical tensions over the deployment of artificial intelligence, U.S. lawmakers are considering how to curb the growing adoption of Chinese AI models by their own companies.
AI has emerged as a key point in the conflict between the US and China, with both countries vying for supremacy in this field.
Chinese models are attracting attention among U.S. companies because they are available at lower prices while closing the performance gap with U.S. rivals.
In April, the Trump administration accused Chinese companies of waging an “industrial-scale campaign” to steal U.S. AI systems and said it would consider ways to hold foreign actors accountable. The Chinese government is considering restricting foreign access to China’s main AI models, Reuters reported on Tuesday.
The increasing adoption of Chinese-made AI models has prompted calls from U.S. lawmakers for strategies to counter this trend, including through ongoing investigations by two U.S. House committees.
“The growing use of Chinese AI models by U.S. companies raises serious concerns,” a State Department spokesperson told CNBC. These “AI models are designed to advance Chinese government discourse, censor dissent, and reflect the ideology and values of the Chinese Communist Party.”
A spokesperson for the British Embassy in the People’s Republic of China said that China “opposes baseless claims and malicious defamation of AI development.” He added that “China’s thriving AI sector is built on independence and strength in science and technology.”
Increased adoption
In April, the House Homeland Security Committee and the House Select Committee on China announced a joint investigation into expanding the use of Chinese-developed AI models. The first step in the investigation was for the chairs of those committees to write to Cursor and Airbnb over their “use of or exposure to these risks” through AI developed in China.
“The Chinese Communist Party is no longer just looming in the artificial intelligence space, but is racing to close gaps in some of the precise capabilities that will shape the future of cybersecurity,” House Homeland Security Committee Chairman Andrew Garbarino told CNBC.
“Recent reports that Chinese promiscuous models rival U.S. leading models in certain vulnerability discovery and cybersecurity tasks are extremely concerning,” Garbarino said.
Although some government departments have banned the use of Chinese AI models, including DeepSeek, U.S. companies are not prohibited from adopting them. Heads of technology, including cryptocurrency companies coinbaseBrian Armstrong and AI startup Lindy’s Flo Crivello have publicly promoted the use of Chinese-made models to reduce costs.
Cursor, which is being acquired by Elon Musk’s SpaceX for $60 billion, built its Composer 2 model using Kimi, a Chinese AI model developed by Moonshot AI. The company declined to comment on the investigation when contacted by CNBC.
Airbnb told CNBC that its “AI activities are overwhelmingly run on models originating from the United States.” The company added that it “uses a limited number of Chinese-originated models, all of which are open source and run only through approved U.S.-based service providers, with data and operations segregated and protected.”
Initiatives for introduction
In addition to focusing on the rise of China’s AI models, the ongoing House Joint Committee investigation is also examining whether the United States is doing enough to respond to the rise of AI models.
“The committee is also examining whether the United States has a sufficient open-weight AI strategy to ensure that U.S. companies and cyber defenders are not forced to choose between expensive or limited U.S.-made models and cheaper, more capable Chinese-developed alternatives,” a committee aide, who requested anonymity because he is not authorized to discuss an ongoing investigation, told CNBC.
Andy Ogles, chairman of the Cybersecurity and Infrastructure Protection Subcommittee, called for a “serious strategy” to ensure the U.S. model is a “true replacement” to the Chinese model.
“If China has a cheaper, more powerful and easier option for AI models, the rest of the world will build on it,” Ogles said in June.
“If we do nothing, the Chinese model will become the default basis for the global digital economy, with embedded censorship, uncertain security, and capabilities extracted from our own laboratories with safety guardrails removed,” he added.
Kyle Zhang, a fellow at the John L. Thornton China Center at the Brookings think tank, told CNBC that the administration could consider using federal procurement bans, including restricting the use of Chinese AI models by government agencies and private companies serving the U.S. government.
“However, it is ultimately impossible to ban China’s open source AI models because the model weights are freely available on the internet,” Zhang added. “This could fall into the First Amendment speech issue.”
While the Trump administration is “clearly concerned” about the risks posed by U.S. companies adopting Chinese AI models, it will be difficult to restrict their use, Daniel Remler, a senior fellow in the Technology and National Security Program at the Center for a New American Security (CNAS) think tank, told CNBC.
Along with potential First Amendment protections, Lemler said the administration may be concerned that “actions against Chinese models could harm startups that use those models or dampen support for open models in general.”
One approach could be procurement requirements that would prevent companies wanting to do business with the government from using Chinese AI models, he added. Another would be to disseminate research findings about the risks and vulnerabilities associated with Chinese AI models to U.S. companies.
“In any case, we hope that both the executive branch and Congress will show interest in ensuring that U.S. companies do not adopt these models,” Lemler said.
