Workflow
全国数据资源摸底:六省份数据产量占近六成,这三个行业最多|言叶知新
Di Yi Cai Jing·2025-05-08 11:40

Core Insights - The report indicates that the data production volume in China reached 41.06 ZB in 2024, marking a 25% year-on-year increase, with expectations to exceed 50 ZB by 2025 [2][8] - The development of artificial intelligence (AI) is significantly driving the demand for data resources, with data used for AI development, training, and inference increasing by 40.95% [3][4] - The eastern region of China is leading in data resource development, with six provinces accounting for nearly 60% of the national data production [8][9] Data Production and Growth - In 2024, the total computing power in China reached 280 EFLOPS, with intelligent computing power accounting for 90 EFLOPS, representing 32% of the total [3][6] - The growth in data production is primarily driven by smart applications, particularly in smart homes and connected vehicles, which saw data growth rates of 51.43% and 29.28%, respectively [3][4] - The number of enterprises utilizing large models increased by over 37%, with the proportion rising to approximately 10% [3] Industry Trends - The manufacturing, finance, and transportation logistics sectors are the top three in terms of data production volume, with finance, mining, and transportation logistics leading in average data production per enterprise [9] - The report highlights that data resource development is concentrated in eastern provinces, with Jiangsu and Guangdong's digital investment being 1.71 times the national average [9] - The demand for high-quality data sets is surging, with the number of high-quality data sets increasing by 27.4% [4][6] Future Outlook - The report anticipates continued growth in data production, with data resource circulation entering an accelerated phase [9] - The integration of AI computing and the rapid enhancement of AI server performance are expected to optimize the structure of data resources [9] - Emerging industries such as autonomous driving and low-altitude economy are projected to maintain high growth rates in data resource scale [9]