China’s Top Tech Giants Begin Shift to Locally Made AI Chips Amid U.S. Restrictions

China’s biggest technology companies are stepping up efforts to replace U.S.-made AI chips with domestically produced alternatives, as ongoing U.S. export restrictions tighten access to Nvidia’s high-performance processors.

Major firms such as Alibaba, Tencent, and Baidu have already started testing and integrating homegrown chips into their systems, signaling a significant pivot toward semiconductor self-reliance. The move is seen as a direct response to U.S. sanctions, which have severely limited the availability of advanced Nvidia chips used for training and running AI models.

While the transition presents technical challenges—including delays of up to three months due to compatibility and performance differences—companies are adopting a hybrid approach. Many continue to use existing Nvidia chips for training complex AI models, while deploying Chinese chips like Huawei’s Ascend for running applications.

To meet the surge in demand, Huawei is ramping up production of its AI chips, though supply still falls short. Other local players, such as Cambricon and Hygon, are also being considered. Some tech firms are even developing their own chips to reduce dependence on foreign technology.

Nvidia has taken a financial hit due to the export curbs, reporting losses of $2.5 billion last quarter and anticipating further declines. CEO Jensen Huang has openly criticized the restrictions, warning they may be accelerating China’s push for chip independence.

As geopolitical and trade tensions persist, China’s tech industry appears determined to secure a future less reliant on U.S. technology, with long-term investments in chip innovation and manufacturing.