Reuters review finds over $3 billion in Chinese AI data-centre projects using Nvidia chips
A Reuters analysis identified more than $3 billion in Chinese computing projects that rely on Nvidia's restricted processors, highlighting the sector's dependence on U.S. technology amid tightening export controls.
A Reuters review has identified more than $3 billion in Chinese computing projects built around Nvidia's restricted processors, even as the American chipmaker says U.S. export controls have effectively shut it out of the AI data-centre market in the world's second-biggest economy. The projects span coastal technology hubs and remote western provinces, underscoring how deeply Nvidia's technology remains embedded in China's AI expansion.
The findings were reported a day after Nvidia co-founder Jensen Huang, who has advocated easing export restrictions, resigned as chief executive amid a scandal. Huang's departure could increase vulnerabilities for Chinese firms that rely on Nvidia hardware, potentially exposing them to delays and supply disruptions as U.S. curbs tighten.
Among the projects identified, a proposed 6.2 billion-yuan (about $873 million) AI computing centre in Fujian province's Xianyou county seeks commercial partners and specifies Nvidia H100 chips for large-scale AI development. In Shaanxi province, a state-owned engineering group’s subsidiary won a contract for a 3 billion-yuan computing centre that calls for Nvidia H100 and H200 processors. A Jilin province government prospectus describes a 2.26 billion-yuan project combining Nvidia H100 chips with domestically developed processors, with bank loans expected to fund roughly 80 % of the investment. In Inner Mongolia, a 1.5 billion-yuan development proposes installing 625 Nvidia H200 servers, and construction of two buildings in its first phase began in June, according to Chinese industry reporting.
Other initiatives have progressed further. In Gansu province, a local company received a 1.16 billion-yuan construction contract in July for a computing centre planning to install about 3,072 Nvidia processors, according to tender documents that list H100 and H200 hardware. The contract was awarded to a consortium led by a subsidiary of state-owned Shaanxi Construction Engineering, though the documents do not confirm delivery of the processors. Existing facilities already use restricted Nvidia hardware: the Wuxi municipal government in Jiangsu province announced in March 2025 that a computing cluster using Nvidia H800 processors had entered operation, and Shanghai-listed Damingcheng disclosed in June that a computing venture in Fujian had built an H800 cluster under a computing-services agreement originally valued at 759 million yuan, which ended early. Smaller purchases indicate demand from research institutions, with Shenzhen University awarding a contract for three H100 processors in September and Northwestern Polytechnical University, which is on a U.S. blacklist, buying two H200 chips the following month.
These plans highlight developers' reliance on U.S. processors that are vulnerable to trade restrictions, and tighter curbs could lead to construction delays, higher costs, or forced redesigns using domestic alternatives.
Reporting by Eduardo Baptista; editing by Shri Navaratnam