SpaceX’s artificial intelligence coding platform, Cursor, is considering abandoning the use of Chinese AI models due to “geopolitical risk,” CNBC has learned.
Cursor, which SpaceX acquired in August for an implied value of $60 billion, had built its Composer 2 coding model on the Kimi K2.5 from Chinese lab Moonshot AI.
It is among U.S. companies that have adopted Chinese models, whose use has increased across the world this year, worrying lawmakers in Washington. The battle for AI supremacy between the United States and China has reached new heights in 2026 as Chinese models have improved.
But now Cursor plans to “explore phasing out open-weight models developed by the PRC (People’s Republic of China) due to geopolitical risk,” according to a report produced by the House Homeland Security Committee and the House Select Committee on China, shared exclusively with CNBC ahead of its release Friday.
The report, which follows a months-long investigation into the adoption of Chinese AI models in the U.S. tech stack, says Cursor “briefed the committees” about these plans. Airbnb and DoorDash also participated in the investigation into their use of Chinese AI models.

SpaceX announced it had entered into a formal agreement to buy Cursor in June; the deal was closed in August. SpaceX is developing the Grok family of models, in which Cursor helped form.
Cursor, SpaceX and DoorDash did not respond to a request for comment from CNBC. Airbnb declined to comment. The U.S. Commerce Department’s Bureau of Industry and Security did not immediately respond to a request for comment.
Chinese AI Adoption
The report finds that the cost, customization, and near-border performance of Chinese AI models are driving adoption, with the United States lacking a competitive and open alternative at the “price and performance demanded by the market.”
“Beijing now seeks to translate these technological advances into global market share and long-term dependence,” the report said. The Chinese embassy in the UK did not respond to a request for comment. China’s Foreign Ministry did not immediately respond.
It also presents recommendations to resolve the problems it has identified.
These include the launch of a “national U.S. open weight initiative” that combines public computing power, shared assessment infrastructure, and “legal access” to cutting-edge U.S. training models.
The report also said it should rely on an order, made public in October, banning DeepSeek from federal civilian agency systems, by prohibiting the use of “PRC-developed AI models” in federal information systems and in the performance of federal contracts.
Separately, Washington is also concerned that Chinese companies are accessing Nvidia chips remotely, through overseas data centers, and gaining ground by using “distillation,” where new models use older, more established models to help train them.
The U.S. Commerce Department’s Bureau of Industry and Security, or BIS, announced an interim final rule on Monday aimed at preventing Chinese companies from remotely accessing AI chips.
Gn bussni
