Data centers full of U.S. chips are serving Chinese AI models–undermining Trump’s tech ambitions
The Nvidia-Megaspeed saga has drawn attention to a significant gap in U.S. export controls: Chinese companies remotely accessing AI chips located outside of China. On Monday, largely in response to the scandal, the U.S. Commerce Department issued a rule closing this loophole.
But while this long-overdue action can help U.S. AI labs widen the gap between their models and those of their Chinese competitors, there’s still a paradox at the heart of Washington’s AI strategy: the U.S. infrastructure spreading around the world is helping Chinese models gain traction.
The Trump administration wants the U.S. to dominate AI from chips to models. Key to the president’s vision is exporting high-end, U.S.-designed chips to countries around the world. Increasingly, these chips—designed by companies like Nvidia—are being deployed in data centers in Latin America, the Middle East, and Southeast Asia.
But U.S. leadership in compute infrastructure doesn’t just help U.S. AI model developers. It helps their Chinese competitors too.
Today, U.S.-owned data centers filled with U.S. chips still rent capacity out to inference providers, which use that capability to serve Chinese large language models (LLMs) to global users.
Chinese models are running on U.S. chips, and the U.S. AI infrastructure build-out is counterintuitively helping Chinese model developers compete. This poses a problem for U.S. AI preeminence, which will need global consumers and enterprises to choose U.S.-developed LLMs over Chinese ones.
The contradiction at the heart of the Trump administration’s dreams of AI dominance is that success at one layer of the stack is undermining its goals at another.
When people hear that Chinese models are gaining users, they generally assume that those models are running on Chinese chips. While that’s increasingly the case within China, the truth is that China cannot produce enough advanced AI chips to satisfy its own companies, let alone meet overseas demand.
Meanwhile, America’s grip on AI infrastructure is formidable. U.S. companies—Nvidia, Google, AMD, and Amazon—design nearly all deployed data-center AI chips. In cloud services, Amazon, Microsoft, and Google together control more than half of the market, with a growing cohort of U.S. “neoclouds” further extending that lead.
Yet the same infrastructure that secures America’s advantage in chips and cloud services gives Chinese AI models a platform to spread. Nvidia itself provides resources to help companies deploy open models—including several Chinese models—on its hardware. U.S.-based model inference providers such as Together AI, Baseten, and Fireworks serve Chinese models to users on infrastructure powered by U.S. chips. For example, DeepSeek’s latest model is being offered by at least 16 inference providers headquartered outside China, including Amazon Bedrock.
When U.S. infrastructure is used to serve Chinese LLMs, it accelerates those models’ global adoption, which in turn benefits China’s soft power and generates revenue for its AI labs. Beijing’s stated ambition is to be the leading provider of transformative AI capabilities, especially in the Global South. Wider adoption of Chinese LLMs advances that objective.
Chinese LLM deployments also generate direct revenue for Chinese model developers. For example, Moonshot AI requires companies that serve its successful Kimi K3 model and exceed $20 million in annual sales to negotiate a revenue-sharing license; Moonshot has reportedly asked Microsoft, Amazon, and Google for 30% of their Kimi K3 revenue.
Today, who benefits from U.S. computing infrastructure depends on whether it serves proprietary or open-weight models. On one side are proprietary LLMs such as Anthropic’s Claude or OpenAI’s GPT. Infrastructure serving proprietary models almost exclusively benefits American developers, because they make the best ones. On the other side are open-weight LLMs, which companies can customize for specialized tasks and run locally. Infrastructure serving open-weight LLMs benefits Chinese developers: users tend to choose the best available models, and in the open-weight market, those are Chinese.
The Trump administration, however, should not go so far as to ban U.S. cloud providers from allowing non-Chinese companies to serve Chinese LLMs. Doing so would incur too high a cost—unjustifiably hindering companies in the United States and allied countries seeking to adopt AI—while failing to deliver sufficient national security benefits. It could also invite Chinese retaliation.
Instead, the United States should compete. Some U.S. companies like Meta and Thinking Machines are already developing open-weight models. If they become more cost-efficient than their Chinese competitors, more U.S. computing infrastructure would serve U.S. models. That shift could limit both the revenue Chinese developers earn from overseas deployments and the soft power China gains from its models’ global reach.
The Trump administration is therefore correct to encourage the development of competitive U.S. open-weight models. It should identify government uses well-suited to open-weight models and direct federal procurement toward American open-weight models for those uses. Finally, as stronger AI regulation becomes necessary, the administration should design those rules carefully to avoid stifling U.S. open-weight model development.
Washington can’t assume that exporting U.S. infrastructure will automatically lead to the export of U.S. models. If it wants to succeed on both fronts, it needs a deliberate strategy for open-weight models.
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
This story was originally featured on Fortune.com
原文: https://fortune.com/2026/11/21/data-centers-us-chips-serve-chinese-models/
