China plans to quadruple its AI computing power by 2030.

Beating AI Insight News Brief: China’s Ministry of Industry and Information Technology (MIIT) has released the "15th Five-Year Plan for the Development of the Information and Communication Industry," setting a national intelligent computing power target of 9,800 EFLOPS by 2030. Latest data from the National Data Administration shows that as of the end of July, China’s total intelligent computing scale reached approximately 2,450 EFLOPS (FP16). Calculated on this basis, the scale needs to expand roughly fourfold over the next four years. The plan also proposes the orderly deployment of 10,000-card, 100,000-card and larger intelligent computing clusters, construction of inference computing power for different scenarios, and increased adaptation of domestic computing chips. To date, China has built 52 intelligent computing facilities with over 10,000 cards. The MIIT disclosed that as of the end of June, the intelligent computing power scale stood at 2,185 EFLOPS, a year-on-year rise of 177%. The plan’s table uses 1,590 EFLOPS in 2025 as the baseline, targeting 9,800 EFLOPS by 2030, equivalent to an approximately 6.2-fold expansion. Meanwhile, the cumulative investment target for information infrastructure across the entire information and communication industry is 3.8 trillion yuan, which covers communication networks and other infrastructure—not all of it is allocated to AI computing power.