House Committees Demand DoorDash Documents Over Use of Chinese AI Model

The House Committee on Homeland Security and the House Select Committee on the Chinese Communist Party have sent DoorDash a letter demanding information on the company's use of Chinese-developed artificial intelligence, according to CNBC, which obtained the letter. The request is part of a joint investigation into the security risks of American companies leaning on AI models built in China.

The letter asks DoorDash to hand over documents related to how it evaluated and deployed AI systems developed in China. Specifically, the committees pointed to a post on X by DoorDash co-founder Andy Fang describing how the company is routing lower-level AI tasks to Kimi K2.6, a model built by Chinese AI firm Moonshot AI.

According to CNBC, DoorDash's AI research lab had said publicly that Kimi K2.6, along with Anthropic's Fable 5, outperformed other Anthropic tools the company had used, including what it called "Sonnet 4.6 and Opus 4.8 harness at a cheaper cost." That public statement is what appears to have put DoorDash on the committees' radar.

DoorDash isn't the first stop on this investigation. The chairmen of the two committees previously sent similar letters to Cursor and Airbnb over their own use of or exposure to Chinese AI models, CNBC reported. DoorDash is the third named company so far.

A DoorDash spokesperson told CNBC the company "proudly supports American AI leadership and is working to ensure AI benefits Main Street, not just the biggest companies." The spokesperson added that DoorDash looks forward to engaging with the committees on how it uses AI "safely and responsibly, including American-developed frontier models and open-weight models."

The Committees' Concern

The letter, as quoted by CNBC, acknowledges a real business logic behind using Chinese open-weight models. They're often cheaper, competitive on performance, and more customizable than relying on a handful of U.S. proprietary providers. The committees didn't pretend otherwise.

But the letter also draws a line: cost savings don't erase the national security question. "Those practical considerations do not eliminate the need for risk-based safeguards or diminish the national security concerns associated with growing dependence on models developed by entities subject to PRC jurisdiction," the letter states, per CNBC.

That's a fair concern on its face. Any AI model developed by a company operating under Chinese Communist Party jurisdiction is, at least in theory, subject to Chinese state influence over data, training, or output, given China's national security laws that can compel domestic companies to cooperate with state intelligence requests. Whether that theoretical exposure translates into actual risk for a U.S. food-delivery company running internal AI tasks is a separate, unresolved question. Nothing in the letter, or in CNBC's reporting, alleges DoorDash user data has been compromised or that any breach has occurred.

Andrew Garbarino, the Republican chairman of the House Committee on Homeland Security, has been vocal about the stakes. "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 told CNBC. He also called reports of a Chinese open-weight model matching U.S. models on cybersecurity vulnerability discovery "highly alarming."

No charges have been filed against DoorDash, and no investigation into wrongdoing by the company itself has been announced. This is an information request, not an accusation. The committees are asking questions, not making findings.

What is established is that DoorDash publicly touted Kimi K2.6's performance and cost advantages over some of its own Anthropic-based tools, according to the X post cited in the letter. That's a business decision DoorDash made in the open, not something uncovered through a leak or subpoena.

The broader pattern is what matters here. Cursor, Airbnb, and now DoorDash have all been pulled into the same congressional inquiry, suggesting this is a systematic look at how deeply Chinese open-weight models like Kimi K2.6 have penetrated mainstream American tech and consumer platforms, not an isolated complaint against one company.

What Happens Next

DoorDash says it intends to engage with the committees. The company has not disclosed a timeline for producing the requested documents, and CNBC's reporting doesn't specify a deadline set by the committees.

The open question is whether this joint investigation results in legislation, such as restrictions on federal contractors or critical infrastructure providers using Chinese-origin models, or whether it stays at the oversight-letter stage. Given that Cursor and Airbnb were contacted first with no publicly reported follow-up enforcement action yet, DoorDash's case will be an early signal of whether these committees have appetite for something with teeth, or whether this remains a documentation-and-disclosure exercise.