Nvidia has reportedly directed its data center partners to reduce their reliance on Chinese customers, according to a Financial Times report. The guidance, communicated verbally through sales teams and regional offices, asks cloud computing providers to ensure that at least 51% of their services are sold to non-Chinese companies. This informal directive applies to new contracts and does not affect existing agreements, though Nvidia has not officially confirmed the policy.
Implications for Chinese AI Firms
The reported requirement would push data center operators to diversify their customer base, potentially limiting the computing capacity available to Chinese businesses. Many Chinese AI companies have relied on overseas cloud providers to access advanced Nvidia processors without directly importing them, circumventing US export controls. If enforced, this could create additional hurdles for these firms, forcing them to seek alternative sources of high-performance computing.
A follow-up report from The Information suggested that Nvidia's instructions are intended to encourage a broader customer mix rather than being a mandatory restriction. This distinction indicates the company is seeking greater oversight over how its chips are used, while avoiding a formal policy that could invite regulatory scrutiny.
Nvidia's Stance on American AI Infrastructure
Nvidia neither confirmed nor denied the report but emphasized the importance of expanding the global reach of American-built AI infrastructure. In a statement cited by AFP, the company said, "Datacenters and clouds built on American chips are the foundation that American model and application makers need to compete worldwide." It added that America's ability to compete in AI depends on its infrastructure being widely adopted across global cloud networks.
Chinese Foreign Ministry spokeswoman Mao Ning told AFP she was unaware of the reported restrictions and expressed hope that companies from all countries would receive a fair, transparent, and non-discriminatory business environment.
Context of US Export Controls
The reported move comes amid Washington's ongoing efforts to restrict exports of advanced AI processors to China, citing national security concerns and potential military applications. Last week, a New York Times investigation revealed that US intelligence officials suspected a Chinese businesswoman associated with Singapore-based data center company Megaspeed of passing commercial information to Beijing. Nvidia CEO Jensen Huang had reportedly developed a close relationship with the executive, though Nvidia has described their interactions as routine business meetings.
Separately, a contractor linked to server maker Super Micro pleaded guilty in October to participating in a scheme to illegally divert Nvidia-powered AI servers to China. Prosecutors allege that two others associated with Super Micro, including a co-founder, conspired to divert approximately $2.5 billion worth of US AI technology to China in violation of export laws.
These developments highlight the increasing scrutiny on AI chip flows and the complexities of balancing commercial interests with national security. For investors, the reported guidance could signal a shift in Nvidia's customer dynamics, potentially affecting revenue streams from Chinese cloud providers. However, the informal nature of the directive leaves room for interpretation and may not lead to immediate changes in existing partnerships.
This article is for informational purposes only and does not constitute financial advice.
