Sou Hu Cai Jing·2025-04-10 06:43Core Insights - The 2025 China Mobile Cloud Intelligence Conference emphasizes the trend of "from cloud to intelligence" driven by the deepening "scale effect" of AI, the accelerating "rebound effect" of computing power, and the expanding "acceleration effect" of connectivity [1][3][6] AI Scale Effects - AI development exhibits two scale effects: the technological scale effect, which leads to an exponential growth in AI capabilities, improving by approximately 10 times annually, and the economic scale effect, which significantly reduces AI usage costs by about 10 times each year [3][4] - By 2030, AI tasks are expected to account for 64% of total network traffic, necessitating urgent innovations in computing network infrastructure [3] Computing Power Rebound Effect - The rebound effect in computing power is expected to result in a more than 2.5 times increase in China's intelligent computing scale over the next three years, with an average annual growth rate of nearly 40% [4] - The demand for inference computing power will surpass that for training computing power, with inference computing expected to grow at nearly four times the rate of training computing over the next three years [4] Connectivity Acceleration Effect - The acceleration of connectivity is fostering new forms of information consumption, leading to the emergence of new business models such as AI smart terminals, intelligent connected vehicles, and intelligent robots [5][6] - By 2030, it is projected that over 80% of new vehicles will be intelligent connected cars, and 80% of humans will interact with intelligent robots daily [5] Structural Transformation of Computing Networks - The convergence of the three effects will drive structural and deep changes in computing network infrastructure, transitioning from traditional cloud computing to a deeply integrated "cloud intelligence computing" model [6][7] - The architecture will expand from IaaS, PaaS, and SaaS to include AI IaaS, AI PaaS, MaaS, and AI SaaS, enhancing resource allocation and AI innovation efficiency [6][8] Strategic Directions for China Mobile - China Mobile aims to strengthen its computing network infrastructure by focusing on three key areas: diversified computing power, integrated computing networks, and comprehensive AI integration [7][9] - The company plans to develop a large-scale "computing power factory," a self-service tool platform, a one-stop model service, and native product applications to meet diverse market needs [8][9] Collaborative Initiatives - China Mobile emphasizes the need for collaboration among various stakeholders to advance cloud intelligence computing, proposing initiatives to tackle core technologies, cultivate inclusive applications, and build a global ecosystem [9]