面向中低收入国家的边缘人工智能洞察报告

The GSMA is a global organisation unifying the mobileecosystem to discover, develop and deliver innovationfoundational to positive business environments andsocietal change. Our vision is to unlock the full power ofconnectivity so that people, industry, and society thrive.Representing mobile operators and organisations across themobile ecosystem and adjacent industries challenges, theGSMA delivers for its members across three broad pillars:Connectivity for Good, Industry Services and Solutions, andOutreach. This activity includes advancing policy, tacklingtoday’s biggest societal issues, underpinning the technologyand interoperability that make mobile work, and providing theworld’s largest platform to convene the mobile ecosystem atthe MWC and M360 series of events. GSMA Intelligence is the definitive source of global mobileoperator data, analysis and forecasts, and publisher ofauthoritative industry reports and research. Our data coversevery operator group, network and MVNO in every countryworldwide – from Afghanistan to Zimbabwe. It is the mostaccurate and complete set of industry metrics available,comprising tens of millions of individual data points, updateddaily. GSMA Intelligence is relied on by leading operators, vendors,regulators, financial institutions and third-party industryplayers, to support strategic decision-making and longterm investment planning. The data is used as an industryreference point and is frequently cited by the media and bythe industry itself. Our team of analysts and experts produceregular thought leading research reports across a range ofindustry topics. We invite you to find out more atwww.gsma.com EmergingTech www.gsmaintelligence.cominfo@gsmaintelligence.com The GSMAEmergingTechprogramme accelerates impact andclimate action by fostering the adoption of AI and emergingtechnologies in low- and middle-income countries (LMICs)by working with public, private and third-sector innovators todevelop scalable and sustainable solutions that have inclusiveand responsible AI at the core. The EmergingTech programmeworks closely with the GSMAAI for Impactinitiative to drivereal-world, impact-focused implementation with telcos inLMICs. Datawise Africa is a research and development companybuilding Africa’s data and intelligent systems, from high-quality datasets to AI models and compute infrastructure.Datawise works across the full data pipeline, researching,building, and deploying solutions designed for real-worldimpact across the continent, partnering with researchers,organisations, and policymakers to shape Africa's digitalfuture. To get in touch with the EmergingTech team, email:emergingtech@gsma.com GSMA Central Insights Unit TheCentral Insights Unit(CIU) sits at the core of GSMAMobile for Development (M4D) and produces in-depthresearch on the role and impact of mobile and digitaltechnologies in advancing sustainable and inclusivedevelopment. The CIU engages with public and privatesector practitioners to generate unique insights and analysison emerging innovations in technology for development.Through our insights, we support international donors to buildexpertise and capacity as they seek to implement digitisationinitiatives in low- and middle-income countries throughpartnerships within the digital ecosystem. Author Eugénie Humeau, GSMA Mobile for Development Contributors Dr. Albert N.Kahira, Datawise AfricaWinny C. Sigilai, Datawise AfricaSylvia Murkomen, Datawise AfricaJames Joiner, GSMA IntelligenceTim Hatt, GSMA Intelligence To get in touch with the Central Insights Unit, email:centralinsights@gsma.com Acknowledgements This report benefited from valuable feedback and insightsfrom colleagues across the GSMA, including Kimberly Brown,Ruth Orbach, Dominica Lindsey, Abi Gleek, Emma Catalfamoand Ambar Khawaja, as well as marketing support fromSimran Jena. We would also like to thank the organisations and expertsconsulted during the research process for sharing theirinsights and perspectives. This includes DEKUT Centre(DSAIL), Prescott Data, Spotlab AI, GIZ, Amini, ITU, AfriclimateAI, Makere University, UET Peshawar, Sophie Bot, ICTP,DataSpires, Jacaranda, University of Dodoma, Sunbird AI,Nelson Mandela IT and Mozilla Foundation. This material is funded by the UK government’s Foreign,Commonwealth & Development Office (FCDO) through itsGlobal Research and Technology Development portfolio. Theviews expressed in this report are those of the authors anddo not necessarily reflect the official policy or positions ofthe UK government’s Foreign, Commonwealth & DevelopmentOffice (FCDO) or its partners. Contents Acronyms and abbreviations2Definitions2List of figures, tables and spotlights4Executive summary61.Introduction81.1.The global AI infrastructure paradigm91.2.The compute gap in LMICs101.3.The opportunity for edge AI111.4.Research objectives and methodology122.Defining compute and edge AI142.1.Compute infrastructure152.2.Edge AI173.Market readiness and unit economics223