China’s Ministry of Industry and Information Technology published a five-year development plan for the information and communications sector that sets a national target of 9,800 exaflops of intelligent computing capacity by 2030. The plan commits ¥3.8 trillion (US$532 billion) in cumulative infrastructure spending across the 2026–2030 period and calls for the deployment of computing clusters equipped with 10,000 or more accelerator cards, positioning the blueprint as Beijing’s most detailed roadmap yet for AI infrastructure at national scale.
Key Takeaways
- The MIIT’s 15th Five-Year Plan for the information and communications industry sets a 2030 target of 9,800 exaflops of intelligent computing capacity, up from 2,185 exaflops recorded at the end of June 2026 and roughly 2,450 exaflops as of July.
- The plan commits ¥3.8 trillion (US$532 billion) in cumulative information infrastructure investment over the five-year period, compared to ¥3.7 trillion in the prior plan cycle.
- New large-scale and hyperscale computing facilities are expected to operate at a power usage effectiveness below 1.2, down from 1.25 in 2025.
- The document calls for 6G commercial services to be initiated “at an appropriate time” and targets 50 5G and 5G-Advanced base stations per 10,000 people, 95% 5G user penetration, and 320 million gigabit broadband subscribers by 2030.
- China has already built 52 intelligent computing facilities equipped with more than 10,000 accelerator cards each, and the plan calls for expanding that buildout to include clusters with 100,000 or more cards.
The Plan Targets a Fourfold Increase in Compute Capacity Against a Backdrop of Rapid Expansion
The document, dated August 12, 2026, and posted publicly through the MIIT’s Information and Communications Development Department on September 7, covers the 2026–2030 period. It contains 13 headline indicators spanning industry development, innovation, green growth, infrastructure, and application adoption, alongside 26 key tasks organized across six priority areas. The intelligent computing target is expressed in exaflops, a unit representing one quintillion floating-point operations per second at FP16 precision.
The baseline figures behind the growth target were detailed at a State Council Information Office press conference on July 20, 2026. Xie Cun, director of the MIIT’s Information and Communications Management Bureau, stated that China’s intelligent computing capacity had reached 2,185 exaflops by the end of June 2026, representing a 177% increase from a year earlier. The overall rack utilization rate across national computing facilities stood at 71.4%. Xie noted that more than 70 high-capacity computing corridors now link national hub nodes, with a roughly 10% improvement in network performance between those hubs over the past two years.
The 9,800-exaflop target would require China to roughly quadruple its current capacity in under four years. The plan pairs the compute target with an advanced storage capacity goal of 1,700 exabytes by 2030, up from 540 exabytes in 2025. Energy efficiency standards also tighten: new large-scale and hyperscale facilities must achieve a power usage effectiveness below 1.2, compared to 1.25 in 2025.
Computing Cluster Deployment Extends the “East Data, West Computing” Architecture
The plan calls for the “orderly deployment” of intelligent computing clusters equipped with 10,000 or more graphics processing cards, as well as clusters using 100,000 or more cards. China already operates 52 facilities at the 10,000-card threshold. The new document signals an acceleration of that buildout and an expansion of individual cluster scale by an order of magnitude.
The infrastructure roadmap builds on China’s existing “East Data, West Computing” network, a national project that routes data processing demand from the densely populated eastern seaboard to cheaper, less congested data center hubs in China’s western and northern provinces. The plan calls for faster construction of a national integrated computing network organized around hub, regional, and edge computing facilities, with improved matching of computing supply and demand across the system.
The document also emphasizes the need to adapt computing infrastructure to domestically produced chips. That directive reflects the ongoing impact of U.S. export controls on advanced semiconductors, which have restricted China’s access to high-end GPUs manufactured by companies outside of China. Rather than pursuing parity at the individual chip level, the plan focuses on a systems-level approach: distributing compute loads across large clusters of domestically available processors, interconnected by high-speed optical networks and coordinated through a national architecture.
The Five-Year Plan Connects AI Compute to a Broader Industrial and Network Strategy
The intelligent computing targets sit within a larger framework that treats AI, 5G connectivity, optical broadband, and data infrastructure as components of a single integrated industrial system. The plan targets information and communications industry revenue of ¥4.1 trillion (US$610.9 billion) by 2030, with average annual growth of 7% in total telecommunications business volume and research and development intensity of 4.8%.
Specific network targets include 50 5G and 5G-Advanced base stations per 10,000 people, a 95% 5G user penetration rate, more than 320 million gigabit-and-above broadband subscribers (up from 240 million in 2025), 6.5 billion communications-network terminal connections, and 500,000 new 5G-Advanced base stations. The plan also calls for the initiation of 6G commercial services “at an appropriate time” during the plan period, alongside the deployment of 400G and 800G optical transport networks and intelligent compute interconnects embedded directly into communications infrastructure.
Wang Zhiqin, vice-president of the China Academy of Information and Communications Technology, described the 2026–2030 period as a “key window” for the country’s information and communications industry to transition from scale leadership to system leadership. Wang stated that the industry needs to coordinate networks, computing power, data, and AI capabilities to support breakthroughs across future industries. The plan also includes measures for supporting the integration of artificial intelligence and manufacturing, building an industry anti-fraud laboratory, and strengthening network and data security governance.
FAQs
What Is China’s Current AI Computing Capacity?
China’s intelligent computing capacity reached 2,185 exaflops by the end of June 2026, up 177% from a year earlier, according to the MIIT. That figure was reported at FP16 precision and represents aggregate capacity across national computing facilities, not a single machine. The country already operates 52 computing facilities equipped with more than 10,000 accelerator cards each.
How Does the ¥3.8 Trillion Investment Figure Compare to Previous Spending?
The ¥3.8 trillion (US$532 billion) cumulative information infrastructure investment target for 2026–2030 represents a modest increase from ¥3.7 trillion in the prior five-year plan period. The figure covers all information and communications infrastructure, not AI computing alone.
What Is the “East Data, West Computing” Network?
The “East Data, West Computing” project is a national infrastructure initiative that routes data processing demand from China’s densely populated eastern provinces to data center hubs in western and northern regions where land, energy, and cooling costs are lower. More than 70 high-capacity computing corridors currently link national hub nodes across the network.
