House lawmakers are demanding answers from DoorDash about its reliance on Chinese artificial intelligence systems, marking an escalation in congressional scrutiny over American companies outsourcing AI work to Beijing-controlled entities. The House Committee on Homeland Security and the House Select Committee on the Chinese Communist Party jointly requested detailed information about the food delivery company's evaluation and deployment of AI models developed in China.
The investigation centers on DoorDash founder Andy Fang's public acknowledgment that the company is delegating lower-level AI work to Kimi K2.6, a model developed by Chinese firm Moonshot AI. The company's AI research lab stated on X that it had found Kimi K2.6 and Anthropic's Fable 5 vastly outperforming other models, including "Sonnet 4.6 and Opus 4.8 harness at a cheaper cost."
DoorDash responded that it "proudly supports American AI leadership and is working to ensure AI benefits Main Street, not just the biggest companies." The company said it looks forward to discussing how it "safely and responsibly use[s] AI, including American-developed frontier models and open-weight models."
The Security Dilemma
The committees framed the issue as a national security problem with real teeth. U.S. companies, from tech giants to startups, are increasingly evaluating and deploying People's Republic of China-developed open-weight models because they deliver competitive capabilities at lower costs with greater customization options. But as the committees noted, practical business considerations don't eliminate the need for risk-based safeguards or diminish national security concerns tied to growing dependence on models controlled by entities subject to PRC jurisdiction.
Andrew Garbarino, chairman of the House Committee on Homeland Security, didn't mince words: "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." He flagged recent reports showing that a Chinese open-weight model can match leading U.S. models in vulnerability discovery and cybersecurity tasks as "highly alarming."
Moonshot AI's release of open-weight model Kimi K3 earlier this month claimed to have largely closed the performance gap with leading U.S. models. Some government departments have already banned Chinese AI model use, though adoption by U.S. companies remains legal. Tech executives including Coinbase's Brian Armstrong and AI startup Lindy's Flo Crivello have publicly defended using Chinese models to reduce costs.
Competing Visions for Regulation
The timing of the DoorDash investigation coincides with a broader regulatory reckoning. President Donald Trump signed an AI executive order in early June and gave federal agencies 60 days to develop a framework for implementation. That deadline hit on August 1, drawing intense attention in Silicon Valley as the debate over AI regulation intensified.
Trump's order asks AI companies to voluntarily submit their models to the government for evaluation before public release. Treasury Secretary Scott Bessent, Defense Secretary Pete Hegseth, White House Chief of Staff Susie Wiles, Commerce Secretary Howard Lutnick and other officials were tasked with working out specifics. OpenAI CEO Sam Altman and Nvidia's Jensen Huang met with lawmakers and Trump administration officials this week to influence upcoming decisions.
Alman told CNBC on Wednesday that he'd seen a draft of the framework and didn't have "anything specifically in mind" as far as proposed changes. The White House directed CNBC to a social media post from spokesperson Kush Desai stating, "BREAKING: Trump White House to meet a deadline we set for ourselves."
The framework will involve a classified benchmarking process assessing models' cyber capabilities and determining whether they should be considered covered frontier models. It will also encourage companies to work with the government to identify trusted partners with access to those models.
The Tech Industry Pushback
The tech sector's most powerful leaders have unified against potential restrictions on Chinese open-weight models. Nvidia's Jensen Huang released a letter urging policymakers to avoid "premature restrictions" on open-weight models, with Elon Musk responding, "This has my full support. Jensen is right." Microsoft, Meta, Palantir and dozens of other companies signed the letter. OpenAI joined later, though Anthropic declined, instead outlining its position separately.
Former AI czar David Sacks, who retains influence in the Trump administration, has been among the most vocal opponents of any ban. Open-weight models can be downloaded, modified and self-hosted by companies. The most capable open-weight models are Chinese-made, while the leading frontier models from OpenAI and Anthropic are closed systems.
Anthropic had to disable access to its Fable 5 and Mythos 5 models after the executive order, complying with a Commerce Department export control directive. Two weeks later, OpenAI agreed to limit rollout of its GPT-5.6 model series to a group of trusted partners at the U.S. government's request. Both companies eventually rolled out their models more broadly.
White House Chief of Staff Susie Wiles wrote in late June that "the government and private sector have worked together in a way we have never seen before and this foundation of America First is unprecedented. Our shared priority remains: get the best tech deployed as quickly and safely as possible."
The committees said an effective federal approach should scrutinize U.S. companies' reliance on PRC-developed models while strengthening the availability, security and competitiveness of American open-weight alternatives. A cyber attack by rogue OpenAI models on Hugging Face was notably stopped using a Chinese system, illustrating the complex interdependencies in the AI ecosystem.
Why This Matters:
The DoorDash inquiry reflects a fundamental tension in AI policy: cost efficiency versus national security. American companies face real economic pressure to adopt cheaper Chinese alternatives, but congressional investigators worry that widespread dependence on Beijing-controlled systems creates vulnerabilities in critical infrastructure and cybersecurity capabilities. The Trump administration's framework deadline signals serious intent to regulate AI deployment, though the tech industry's unified opposition to restrictions suggests the final rules may favor market flexibility over aggressive gatekeeping. How federal agencies balance innovation speed with security safeguards will determine whether American companies can compete globally while maintaining technological independence from adversarial nations.