China’s leading tech companies are turning to homegrown chips after the United States tightened export rules on advanced processors used for artificial intelligence. The shift comes after the US banned the export of Nvidia’s powerful H20 chip, widely used in AI systems, to China. This move has forced companies like Alibaba, Tencent, and Baidu to search for local alternatives to keep up with AI development.
According to the Financial Times, China’s supply of Nvidia’s H20 chips will likely run out by early next year. Even if Nvidia creates a version that meets US restrictions, the delivery process may take up to six months. In addition, the new chips may lack high-bandwidth memory, which is essential for processing large amounts of data.
Nvidia has stated that they are working on export-compliant chips, expected to be ready by July. However, the company has not confirmed the full technical details. For now, Chinese companies must plan ahead to avoid disruptions in AI projects.
In response, companies such as Baidu and Alibaba are testing a range of domestic chip options. They are confident that Chinese-made chips, combined with local software solutions, can help meet rising demand for AI technology. These efforts reflect a broader move by Chinese tech leaders to build independent technology systems and reduce reliance on US products.
One notable alternative is Huawei’s Ascend chip series, which has recently seen increased interest. Reports from Chinese research institutes suggest that companies have begun large-scale purchases of Ascend chips. These chips are designed in China and optimized for AI workloads. Despite international warnings about the legal risks of using Huawei technology, the chips are becoming a key part of China’s AI plans.
Switching from Nvidia chips to domestic ones is not simple. It requires time, investment, and technical support. Companies may need to rebuild their systems and train engineers to work with new chip architectures. Migrating to new platforms also involves software adjustments, which can delay progress in AI research and product development.
In addition to using Huawei chips, some companies are also designing their own processors. Alibaba and Baidu are among those developing in-house chip designs. This gives them more control over performance and supply, reducing the impact of foreign restrictions. These self-developed chips are part of a long-term strategy to build a fully independent AI ecosystem.
China’s ambitions extend beyond replacing Nvidia chips. The country wants full control over its entire AI system—from hardware and software to cloud platforms and data centers. China is even planning to expand its AI infrastructure into space, aiming to use satellites and other orbital systems to power advanced computing. This reflects a strong national drive to lead in global technology development.
While the US export ban aimed to slow China’s progress, it may have pushed Chinese companies to innovate faster. By investing in local solutions, China could reduce its dependence on foreign technology and gain an edge in the global AI race. These changes mark a major shift in the tech world, as China builds its own path forward in artificial intelligence.
