Chinese AI Firms Seek Overseas Nvidia Computing Capacity Amid Fears of Tighter US Chip Controls

Chinese artificial intelligence companies are seeking to secure access to Nvidia-powered computing capacity at overseas data centers as concerns grow that Washington could tighten restrictions on the remote use of advanced processors by Chinese customers.

The moves reflect growing uncertainty among Chinese technology companies over whether they will continue to be able to access high-performance computing resources located outside mainland China. Although the United States restricts direct exports of some advanced Nvidia chips to China, companies can currently obtain computing power remotely by renting capacity at data centers in other countries.

According to an exclusive report by Reuters, a major Chinese cloud provider has discussed internally whether to reserve overseas Nvidia computing capacity, while several smaller AI developers have approached data-center operators about securing access. Some companies are reportedly considering providers in Southeast Asia.

The concern is that Washington could close what Chinese companies view as a gap in the existing export-control framework. The proposed Remote Access Security Act, which has passed the U.S. House of Representatives and is awaiting further action in the Senate, would give the U.S. Commerce Department greater authority to regulate remote access to restricted technology.

Advanced graphics processing units are essential for training and operating large artificial intelligence models. They are used in applications ranging from generative AI chatbots and image-generation systems to industrial automation, scientific research and advanced data analysis.

For Chinese developers, access to powerful processors remains a major factor in competing with international AI companies. Restrictions on direct chip imports have already pushed some firms to explore alternative sources of computing power, including domestically developed processors and overseas cloud services.

The latest developments have been complicated by controversy surrounding Nvidia Chief Executive Jensen Huang and Alice Huang, a former executive at Singapore-based cloud-computing company Megaspeed. U.S. authorities reportedly suspected Alice Huang of having links to Chinese intelligence, an allegation that has not been independently verified and has been denied by China. Megaspeed has denied wrongdoing and said it is cooperating with U.S. scrutiny.

Reuters reported that the controversy has increased concern among Chinese AI companies that political pressure could accelerate efforts to regulate overseas access to restricted processors. The report also said the situation had contributed to renewed buying interest in China's grey market for advanced Nvidia chips.

The issue is particularly significant because the global AI industry depends heavily on a relatively small number of advanced semiconductor suppliers. Nvidia's high-performance processors are widely used in large-scale AI training, and restrictions on access could affect the speed and cost of development for companies that rely on them.

Chinese firms are attempting to reduce their dependence on foreign technology by investing in domestic chip design, semiconductor manufacturing and alternative AI computing platforms. However, replacing the performance, software ecosystem and scale of Nvidia's products remains a challenge for many developers.

The potential expansion of U.S. restrictions could also affect data-center operators in countries that provide cloud computing services to international customers. Providers may face additional compliance requirements if Washington seeks to monitor who is using restricted chips, where users are located and how computing resources are deployed.

For Southeast Asian economies, the issue could create both commercial opportunities and regulatory challenges. Data-center operators may benefit from additional demand from Chinese clients, but they could also face pressure to demonstrate compliance with U.S. export-control rules.

The developments underline how semiconductor restrictions are increasingly extending beyond the physical movement of chips to include access to computing services. Whether the United States ultimately adopts new rules will be important for the future of cross-border AI development and the competitive position of Chinese technology companies.