Nvidia faces competing US and Chinese chip restrictions

Nvidia’s processors and CUDA software ecosystem have made the company central to the contest for AI leadership between Washington and Beijing. Its GPUs support demanding workloads, including model training, scientific research and military applications, while its established software tools make switching suppliers costly and complicated.

The US has tightened advanced semiconductor export restrictions since October 2022 to limit China’s access to computing power with potential military uses. In January 2026, the Commerce Department introduced case-by-case licensing for Nvidia’s H200 processors and comparable chips, subject to supply, compliance and testing conditions. Beijing reportedly restricted purchases to special circumstances while encouraging domestic alternatives. Around 600,000 H200 processors were reportedly awaiting mainland shipment in Hong Kong.

China remains an important market despite growing competition. IDC data reported by Reuters put Nvidia’s share of China’s data-centre AI chip market at 34% in the first half of 2026, compared with Huawei’s 28%. Chinese developers continue using advanced Nvidia hardware, while access through overseas data centres complicates enforcement of US restrictions.

Nvidia also faces scrutiny surrounding Megaspeed, a Singapore-based company investigated over suspected chip diversion to China. Nvidia’s board has launched an independent investigation into whether chief executive Jensen Huang was targeted by foreign intelligence. The alleged intelligence links remain unproven, and Nvidia says it has no reason to believe company secrets or US national security were compromised. US lawmakers are seeking answers and stronger export oversight.