人工智能（AI）赋能政府：数字先锋报告

A digital sprinters report Contents Executive summary 3 Introduction6 The AI-driven socioeconomic opportunity for governments7 Four country archetypes: Benchmarking AI readiness19 AI in action: Public sector AI use case strategy23 AI skills: Equipping governments for the AI era29 Recap and conclusion 35 Acknowledgments39 Appendix40 Executive summary The opportunity presented by artificial intelligence (AI) is particularly profound for governments in emerging markets.Here demographic momentum, rapid digital adoption, and the absence of legacy constraints converge to create a uniqueposition, enabling these markets to leapfrog traditional development pathways — if emerging market governmentleaders act decisively and strategically. Building on this foundation, this report provides, for the first time, a quantification of AI’s public-sector-specific economicpotential in emerging markets. These gains are grounded in dynamic modeling that quantifies both direct productivityimprovements and the wider spillovers generated by government AI adoption. AI can help deliver value for governments through three main pathways. 1.Government efficiency and fiscal health:Improving how resources are raised, allocated and spent; reducing fraudleakage, and duplication; and strengthening long-term fiscal sustainability. 2.Public good and service delivery:Enhancing the reach, quality and fairness of public services in areas such ashealth, education, infrastructure and citizen engagement, while building public trust and institutional resilience. 3.Economic growth and prosperity:Contributing directly to GDP growth, productivity and household income bydriving efficiency, creating new jobs and leading by example private-sector adoption. Our modeling suggests that widespread AI adoption within thepublic sector could increase public administration productivity byup to 3%. This would increase real gross domestic product (GDP)by up to 4%, reduce government deficit by as much as 22%,reduce unemployment by as much as 1.5 percentage points, andincrease household incomes by 2% by 2035 when compared tothe baseline scenario in 2035. 4%22%gross domesticproductgovernmentfiscal deficit These benefits of AI in the public sector are deeply interconnected and mutually reinforcing.A more capable taxauthority funds better infrastructure. Streamlined licensing encourages small business growth. Improved agriculturalextension services connect farmers with information and technology, which can boost rural incomes. Smarter healthcaresystems keep citizens healthier and more productive. Together, AI-supported services can help create a more efficient,adaptive, equitable, and trusted government—driving further downstream adoption and compounding benefits. The realization of these benefits depends on several key enablersas outlined inGoogle’s AI Sprinters framework. •A cloud-first, AI-forward policy:Reliable access to secure,scalable computing resources—whether through regionalor global public cloud—accelerates time-to-value andencourages private investment. •Skilling at scale:Broad-based literacy, implementer capacityand selective innovator pipelines should be developed in linewith country realities. •Modern data systems:High-quality, representative datasetsand trusted data flows are essential for relevant, unbiased AIsolutions. Explorers Early in adoption; building national AI strategies andfoundational mobile-first AI services. •Enabling policies:A clear national strategy, centralcoordination and continuous feedback loops are necessary totrack progress and build public trust. Countries early in AI adoption, typically without cloudfirst policies and a national AI strategy; need to focus onestablishing these foundations and closing gaps in basicinfrastructure and digital capabilities. Our research identified four archetypes of AI readiness.Thisreport introduces a four-archetype framework that benchmarkspublic sector AI readiness in emerging markets which can beused to guide the sequencing of investments and use cases. Infra Ready This framework is a self assessment starting point—not aranking—to help governments prioritize feasible use cases nowwhile laying the groundwork for longer-term scaling. Strong connectivity; developing governance capacity;targeting sector-specific AI use cases. Countries with strong connectivity and cloud infrastructurebut limited institutional readiness; well placed to deployoperations focused and agentic solutions while buildingcohesive AI governance and a cross ministry task force. Effective implementation of these enablers has already shownthat AI can be transformative across multiple governmentdomains. Here are a few notable public sector use cases. •Public finance and revenue:Fraud detection, smarteraudits and procurement analytics strengthenfiscal stewardship. Governance Ready Infrastructure limited but strong on policy andoversight frameworks; leveraging public cloud. •Citizen services and