WASHINGTON, November 18, 2026, The Trump administration is considering whether to require US companies to disclose publicly their use of artificial intelligence models developed in China, according to Reuters, which cited three people familiar with the discussions. No rule has been announced or adopted.
A possible transparency rule, not a prohibition
The Treasury Department and the White House Office of Science and Technology Policy have discussed a plan under which affected businesses would post information about their Chinese AI use on their websites, Reuters reported. Rather than blocking the technology outright, such a policy would seek to discourage dependence through public scrutiny.
The report explicitly said it was unclear whether the proposal was gaining support or would proceed to formal rulemaking. Neither Treasury nor the White House responded to Reuters' requests for comment. As of the report's publication on November 18, the deliberations were not an enforceable legal requirement.
The distinction between an internal discussion and a final rule is consequential for businesses. Companies should not treat the reported proposal as an immediate compliance deadline, nor assume that a final version would necessarily cover every kind of AI application.
Why US businesses are using Chinese models
Open-weight AI models make key model components available for download and adaptation. Reuters reported that Chinese offerings are attracting US users because they can cost less than proprietary products from American laboratories while offering useful performance.
Airbnb is among the examples identified by Reuters: the company uses Alibaba's Qwen technology in an AI-powered customer-service agent. Airbnb said its AI activity runs overwhelmingly on US-origin models and that its limited use of China-origin open-source models is through approved US-based service providers, with data and operations protected.
That response illustrates a central technical distinction. A model's country of origin is not the same thing as the physical location where it runs. Companies may host downloadable models within their own systems or use third-party services. The security and data-governance implications depend on the actual deployment and controls, not simply the developer's nationality.
Reuters also identified congressional interest in the issue. Representatives Andrew Garbarino and John Moolenaar, who lead House committees dealing with homeland security and China respectively, launched an investigation this year into American companies' use of Chinese AI. Their inquiries included letters to Airbnb, Anysphere, the maker of Cursor, and DoorDash.
Moolenaar welcomed the reported disclosure discussions while arguing that risks could remain even when Chinese-origin models operate on US infrastructure.
A dispute over technological dependence
Some Washington policymakers fear that increasingly capable Chinese open-weight models could make US businesses dependent on foreign-developed technology. China rejects efforts to frame ordinary technological competition as a security threat. Its embassy in Washington told Reuters that politicising technology and trade would impede AI progress and said Beijing would defend Chinese companies' legitimate interests.
Those competing positions do not establish that every Chinese model is unsafe, or that every use of one creates a data-transfer risk. The report documents a policy debate and corporate responses, not a finding of wrongdoing by the businesses named.
What remains unresolved
The proposed disclosure's scope, legal authority, implementation timetable and enforcement mechanism have not been specified publicly in the Reuters report. It is also unclear whether the administration would pursue a formal rule or abandon the idea.
If adopted, a disclosure obligation could influence corporate procurement and encourage businesses to review which models power customer-facing and internal systems. But that is a potential consequence, not an enacted policy. The immediate development is that US officials have reportedly discussed a transparency measure amid broader competition over AI technology.




