Zhong Guo Jing Ying Bao·2025-04-16 11:20Core Insights - China Mobile has achieved annual revenue exceeding 1 trillion yuan and is advancing its cloud computing strategy for 2025, focusing on a comprehensive computing network that integrates various types of computing power [1] - The chairman of China Mobile, Yang Jie, highlighted two significant changes driven by innovations in general artificial intelligence: exponential growth in total computing power and a shift in computing power structure, with inference computing demand expected to surpass training computing demand in the coming years [1][2] Group 1: Computing Power Network - China Mobile has established the first nationwide computing power network that integrates general computing power (8.5 EFLOPS), intelligent computing power (43 EFLOPS), quantum computing power (1138 Qubit), and supercomputing power (800 PFLOPS), accounting for one-sixth of the national total [1] - The intelligent computing power scale is projected to grow over 2.5 times in the next three years, with an annual compound growth rate of nearly 40% [1] Group 2: AI Development Effects - Yang Jie identified two scale effects in AI development: a technological capability scale effect leading to a tenfold increase in AI capabilities annually, and an economic benefit scale effect that will reduce AI usage costs by approximately tenfold each year [2] - By 2030, AI tasks are expected to account for 64% of total network traffic, necessitating urgent innovations in computing network infrastructure [2] Group 3: Next-Generation Cloud Computing Paradigm - China Mobile released the "Cloud Intelligence Computing Technology White Paper," proposing a next-generation cloud computing paradigm that integrates cloud and AI [3] - The new architecture will evolve from traditional IaaS, PaaS, and SaaS to AI IaaS, AI PaaS, MaaS (Model as a Service), and AI SaaS, facilitating a transition from "computing network brain" to "intelligent computing network brain" [3][4] Group 4: Strategic Initiatives - China Mobile announced four strategic initiatives: building an integrated computing network infrastructure (AI IaaS), creating a self-service tool platform (AI PaaS), developing a one-stop model service (MaaS), and launching native product applications (AI SaaS) [6] - These initiatives aim to enhance AI innovation efficiency and support the comprehensive digital transformation of production, lifestyle, and social governance [4][6] Group 5: Emerging Business Models - The new strategy indicates the emergence of new consumption patterns in the cloud industry, particularly through the integration of AI smart terminals with cloud computing [7] - The launch of the mobile AI cloud computer, which integrates advanced GPU technology, aims to enhance user experience by significantly improving performance metrics such as CPU load reduction and graphics rendering capabilities [8] Group 6: Industry Collaborations and Future Focus - China Mobile, in collaboration with companies like Moore Threads, initiated the "Intelligent Computing Interconnection OISA Industry Chain Initiative" to enhance high-performance GPU interconnection technology [9] - The focus on smart connected vehicles is expected to grow, with projections indicating that by 2030, over 80% of new cars will be smart connected, supported by high-performance computing and low-latency networks [9]