U.S. companies are increasingly turning to Chinese artificial intelligence models to cut costs, and Congress wants answers. The House Committee on Homeland Security and the House Select Committee on China launched a joint investigation three months ago into the growing reliance on Beijing-developed AI systems, sending formal inquiries to companies including Cursor and Airbnb about their exposure to these tools.
The economic incentive is straightforward. Chinese AI models are closing the performance gap with American competitors while costing significantly less to deploy. Tech leaders from Coinbase to Lindy have publicly championed Chinese models as a cost-effective alternative. Cursor, soon to be acquired by Elon Musk's SpaceX for $60 billion, built its Composer 2 model using Kimi, a Chinese AI system developed by Moonshot AI.
The Security Concern
U.S. officials view this trend as a national security problem. Andrew Garbarino, chairman of the House Committee on Homeland Security, told CNBC that Chinese capabilities in AI have advanced dramatically. "The Chinese Communist Party is no longer just nipping at our heels in artificial intelligence; it is racing to close the gap in some of the exact capabilities that will shape the future of cybersecurity," Garbarino said. He flagged recent reports showing Chinese open-weight models can match leading U.S. systems in vulnerability discovery and cybersecurity tasks.
The State Department expressed similar alarm. "The growing use of Chinese AI models by U.S. companies raises serious concerns," a State Department spokesperson told CNBC. The agency contends these models "are designed to advance Beijing's narratives, censor dissent, and reflect CCP ideology and values."
China's response dismissed the criticism. A spokesperson for the U.K. embassy of the People's Republic of China said the country "opposes baseless allegations and malicious smears against its AI development," characterizing China's AI sector as "built on self-reliance and strength in science and technology."
The Policy Dilemma
Some government departments have already banned Chinese AI models including DeepSeek. But U.S. companies face no such restriction. The committees are investigating whether the government is doing enough to address the trend and whether the U.S. has developed a sufficient strategy around open-weight AI to give American firms a real choice beyond expensive or restricted domestic options versus cheap, capable Chinese alternatives.
Andy Ogles, chairman of the Subcommittee on Cybersecurity and Infrastructure Protection, outlined the stakes plainly one month ago. "When the cheap, capable, easy option for an AI model is Chinese, the rest of the world will build on it," Ogles said. "If we do nothing, Chinese models become the default foundation of the global digital economy, carrying embedded censorship, uncertain security, and capabilities distilled from our own laboratories with the safety guardrails stripped out."
Policymakers are considering several approaches. Kyle Chan, a fellow in the John L. Thornton China Center at Brookings, told CNBC that federal procurement bans could restrict government agencies and private contractors from using Chinese models. But he acknowledged a fundamental problem: "It's ultimately impossible to ban China's open-source AI models because their model weights are available freely on the internet." Such restrictions could also raise First Amendment questions.
Daniel Remler, senior fellow in the technology and national security program at the Center for a New American Security, suggested the administration faces a balancing act. The Trump administration is "clearly worried" about risks from American adoption of Chinese AI, Remler told CNBC, but heavy-handed restrictions risk harming startups that depend on these models or chilling broader support for open-source development.
Remler outlined potential middle-ground approaches: procurement requirements that discourage government contractors from using Chinese models, or public disclosure of vulnerabilities and risks associated with these systems to inform company decision-making. "Regardless, I do expect both the Executive Branch and Congress to communicate their interest not to see U.S. companies adopting these models," Remler said.
Airbnb defended its limited use of Chinese AI. The company told CNBC that its "AI activity runs overwhelmingly on U.S.-origin models" and that it uses only a "limited number of China-origin models, all of which are open-source and run only through approved U.S.-based service providers, keeping data and operations separate and protected." Cursor declined to comment on the investigation.
Why This Matters:
The investigation reveals a core tension in modern industrial policy: market forces are pushing U.S. companies toward cheaper Chinese AI tools, but national security officials worry about the long-term consequences of ceding technological dominance to a strategic competitor. The challenge for policymakers is substantial. Outright bans on open-source software raise constitutional and practical obstacles, yet allowing unfettered adoption of Chinese models could embed Beijing's technology into America's digital infrastructure. The outcome will likely shape whether the U.S. maintains competitive advantage in AI development or becomes dependent on systems designed by and potentially vulnerable to foreign control. How Congress and the administration balance innovation incentives against security concerns will determine whether American companies remain leaders in this critical technology or become followers using someone else's tools.