GTHT: Domestic Intelligent Computing Accelerates Deployment as National Computing Power Demand Surges

Stock News
Aug 27

Guotai Haitong Securities Co., Ltd. released a research report stating that on August 23, 2025, MIIT Deputy Minister Xiong Jijun announced at the 2025 China Computing Power Conference that the ministry will (1) improve the computing power layout policy system, and (2) optimize computing power infrastructure layout, guiding regions to reasonably deploy intelligent computing facilities. On the same day, the China Computing Power Platform was officially connected, with the first batch including 10 provinces and cities such as Shanxi, Zhejiang, and Shanghai. The firm believes these measures are expected to coordinate the national computing power center layout, accelerate local intelligent computing center construction, and drive growth in demand for domestic computing power chips.

GTHT's main viewpoints are as follows:

**MIIT Optimizes Computing Power Infrastructure Layout, Intelligent Computing Center Construction Expected to Accelerate**

On August 25, 2025, MIIT Deputy Minister Xiong Jijun stated at the 2025 China Computing Power Conference that as of the end of June 2025, China's operational computing power centers have reached 10.85 million standard racks, with intelligent computing power scale reaching 788 EFLOPS. Going forward, MIIT will (1) improve the computing power layout policy system, and (2) optimize computing power infrastructure layout, guiding regions to reasonably deploy intelligent computing facilities. The firm believes these measures will effectively address issues such as "whether local intelligent computing centers can be built" and "how to build them," expected to accelerate local intelligent computing center construction and drive growth in demand for domestic computing power chips.

**Integrated Computing Network Construction Accelerates, China Computing Power Platform Officially Connected**

The "Computing Power Interconnection Access Ceremony" between the National Supercomputing Internet and seven city computing power centers was successfully held at the main forum of the 2025 China Computing Power Conference. With the official access of seven computing power centers including the National Supercomputing Taiyuan Center and Qingdao's "Heart of the Sea" Artificial Intelligence Computing Center, the backbone node-type computing power centers connected to the National Supercomputing Internet have exceeded 30. Meanwhile, the China Computing Power Platform was officially connected, with the first batch including 10 provinces and cities such as Shanxi, Zhejiang, and Shanghai. The firm believes the connection of the China Computing Power Platform and integrated computing network construction are expected to coordinate the national computing power center layout, accelerate domestic AI industry development, and domestic computing power demand is expected to benefit.

**Catalyst:** Accelerated growth in local intelligent computing center investment.

**Risk Warning:** Domestic computing power chip supply capacity falls short of expectations; domestic AI large model progress falls short of expectations.

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10