China Proposes $295 Billion AI Data Center Grid Focused on Domestic Chips

China is reportedly drafting an ambitious plan to establish a nationwide network of AI data centers, backed by an investment of around 2 trillion yuan ($295 billion) over the next five years. The core objective of this initiative is to ensure that a substantial majority, specifically 80%, of the technology employed within this infrastructure, including AI processors, originates from Chinese companies.

Sources suggest that major state-owned telecommunications providers, China Mobile and China Telecom, will be responsible for operating the majority of these facilities. The aim is to connect them into a singular, powerful computing grid by the year 2028. This substantial infrastructure build-out is expected to be financed through sovereign debt and specialized long-term government bonds. Some estimates indicate that including necessary power grid upgrades could push the total capital expenditure beyond 5 trillion yuan.

However, the project faces significant hurdles, primarily concerning the domestic supply of advanced semiconductor technology. The mandate for 80% domestic sourcing effectively excludes major international chipmakers like Nvidia and AMD. China's reliance will fall on local foundries such as SMIC, whose most advanced available manufacturing process is reportedly equivalent to 7nm and is already operating at very high utilization rates. This limited foundry capacity, coupled with restricted domestic production of high-bandwidth memory (HBM), could bottleneck the scaling of Chinese AI accelerators, such as those developed by Huawei.