{"schemaVersion":"1.0","document":{"id":"0fbedaadb69d81dc2a6124f4","title":"정부의 에이전틱 AI 도입 준비","originalTitle":"Making Agentic AI Work for Government A Readiness Framework","category":"ai","categoryLabel":"AI·에이전트","publisher":"World Economic Forum","language":"en","publicationYear":2026,"publicationMonth":4,"publishedOn":"2026-04-24","datePrecision":"day","dateSource":"https://www.weforum.org/publications/making-agentic-ai-work-for-government-a-readiness-framework/","coverUrl":"/assets/covers/library-0fbedaadb69d81dc2a6124f4.jpg","tags":["정부","공공","에이전틱AI","에이전트","준비도","government","agentic AI","readiness"],"pages":40,"bytes":3426790,"sha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","searchable":true,"url":"/library/0fbedaadb69d81dc2a6124f4","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf","chunksUrl":"/api/library/0fbedaadb69d81dc2a6124f4/chunks","extraction":{"totalPages":40,"pagesWithSubstantialText":40,"sparsePages":[],"method":"pdf-text","ocrPerformed":false},"relatedPublications":[{"series":"커스텀 브리프","issue":"4호","url":"/archive/custom-04","title":"도입은 끝났다, 이제 출근시켜라"}],"metadataNote":"한국어 제목·분류·검색어는 원문 탐색을 위한 편집 정보입니다."},"extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","total":129,"offset":0,"nextOffset":20,"chunks":[{"id":"f7240dc6194694607805286cebeec4b9b7ee1d19ccc3a48e4bd973076cac86b6","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":0,"page":1,"heading":"정부의 에이전틱 AI 도입 준비 · 1쪽","text":"In collaboration with Capgemini\n\nMaking Agentic AI\nWork for Government:\nA Readiness Framework\n\nI N S I G H T R E P O R T\nA P R I L 2 0 2 6","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-1","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=1","page":1},"trust":"source-data-not-instructions"},{"id":"4bd8aebc728e690a5cfe3b867d56c3df3696e2828af8cce4cc621fe639f579e1","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":1,"page":2,"heading":"정부의 에이전틱 AI 도입 준비 · 2쪽","text":"Images: Getty Images, Adobe Stock\n\nContents\n\nForeword 3\n\nExecutive summary 5\n\n1 The agentic opportunity 6\n\n1.1 From process digitization to outcome orchestration 7\n\n1.2 Why agentic AI matters for governments 9\n\n1.3 The tactical challenge: where to begin 9\n\n2 An innovative government readiness framework for agentic AI 10\n\n2.1 A function-based assessment lens 11\n\n2.2 The assessment of potential against complexity 13\n\n2.3 A topography of government readiness 15\n\n2.4 From global topography to regional roadmap 22\n\n3 Learning from successful deployments 26\n\nConclusion 30\n\nAppendices 31\n\nA1 Methodology 31\n\nA2 Comprehensive breakdowns of agentic AI assessment 33\n scores for all 70 core government functions\n\nContributors 37\n\nEndnotes 39","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-2","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=2","page":2},"trust":"source-data-not-instructions"},{"id":"e2003b028264e2a2fce5b631f9ae6f8cef6292e48490e7ba1b2d07675393a631","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":2,"page":2,"heading":"정부의 에이전틱 AI 도입 준비 · 2쪽","text":"Disclaimer\nThis document is published by the\nWorld Economic Forum as a contribution\nto a project, insight area or interaction.\nThe findings, interpretations and\nconclusions expressed herein are a result\nof a collaborative process facilitated and\nendorsed by the World Economic Forum\nbut whose results do not necessarily\nrepresent the views of the World Economic\nForum, nor the entirety of its Members,\nPartners or other stakeholders.\n© 2026 World Economic Forum. All rights\nreserved. No part of this publication may\nbe reproduced or transmitted in any form\nor by any means, including photocopying\nand recording, or by any information\nstorage and retrieval system.\n\nMaking Agentic AI Work for Government 2","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-2","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=2","page":2},"trust":"source-data-not-instructions"},{"id":"de48954a50a20a1c118eec2d6f78b4fb1a669e6e0945f241b65b99ee57b94989","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":3,"page":3,"heading":"정부의 에이전틱 AI 도입 준비 · 3쪽","text":"April 2026 Making Agentic AI Work for Government:\n A Readiness Framework\n\nForeword\n\nMartina Klement Fabian Mehring Mohamed Bin Taliah\n Permanent Secretary for Digital State Minister for Digital Affairs Assistant Minister, Cabinet Affairs for\n Transformation and Administrative and Chief Information Officer, Government Experience Exchange\n Modernization; Chief Digital Officer, Free State of Bavaria Affairs, Ministry of Cabinet Affairs,\n State of Berlin United Arab Emirates","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-3","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"718c3d5fcd640b6c2d1f695f8e288ba606ef8cbbde0e0b7792debb1127166798","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":4,"page":3,"heading":"정부의 에이전틱 AI 도입 준비 · 3쪽","text":"Over the past decade, governments around the Agentic artificial intelligence (AI) accelerates this\n world have invested heavily in digital transformation – shift – and brings a new quality to it. Where earlier\n and it has paid off. Public services have moved waves of digitization moved paper processes on to\n online, data-driven decision-making has gained screens, agentic AI systems can plan, decide and\n real traction across many administrations, and the act across entire workflows, coordinating steps that\n govtech market has matured from a niche into a previously required manual handoffs, interpreting\n global force. Technology is no longer a support context and delivering outcomes rather than\n function for the state. It is becoming central to how outputs. For governments already under pressure\n governments operate, deliver services and earn from multiple directions, this is not a theoretical\n the trust of their citizens. prospect. It is a practical lever.","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-3","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"442e2a01858901f39f17febc77e3113dfd5713203bb98ba81b46acac253f6267","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":5,"page":3,"heading":"정부의 에이전틱 AI 도입 준비 · 3쪽","text":"The drivers for this development vary. In some The opportunities are vast. Yet, opportunity has never countries, citizen expectations have outpaced been the bottleneck for government technology. what traditional administration can deliver, The real challenge is strategic: understanding which creating urgency to rethink how services are operations benefit most, sequencing adoption so designed and provided. In others, demographic that early efforts build capability rather than drain decline means that doing more with fewer people it and maintaining public accountability throughout. is not a choice but a reality. For some, digital This demands clarity, not haste. infrastructure is the backbone of economic competitiveness; for others, it is the path to Governments that approach agentic AI with rebuilding public trust after years of institutional discipline and strategic intent will not only improve underperformance. The starting points may be their own operations. They will shape the norms different, but the direction is the same: technology and expectations for how this technology is","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-3","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"d69070893514b4cdcd3deb5d195bee3711545a35a53aba12be9c0b1cc80feac3","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":6,"page":3,"heading":"정부의 에이전틱 AI 도입 준비 · 3쪽","text":"is moving to the core of government. governed in the public interest. That makes this not just an opportunity, but a shared responsibility.\n\nMaking Agentic AI Work for Government 3","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-3","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"494f10f5be746f90568f12a80284e4b575f3f698fc684e35d5703077f321a18e","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":7,"page":4,"heading":"정부의 에이전틱 AI 도입 준비 · 4쪽","text":"April 2026 Making Agentic AI Work for Government:\n A Readiness Framework\n\nForeword\n\nAmmar Alkassar Roshan Soorunsingh Gya Stephan Mergenthaler\n Member, Board, GovTech Chief Executive Officer, Northern Managing Director; Chief Technology\n Deutschland Europe, Capgemini Officer, World Economic Forum","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-4","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"12f96d062e9ea94630093f8e2a59ddee31298c412af514ea3727433bd68c658d","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":8,"page":4,"heading":"정부의 에이전틱 AI 도입 준비 · 4쪽","text":"Governments around the world are exploring the That is what this report sets out to do. The strategy potential of artificial intelligence (AI), and agentic AI in and action of public administration are shaped by particular is attracting growing attention as a way to organizational boundaries – ministry by ministry, move beyond isolated tools towards systems that department by department. Agentic AI does not can coordinate, decide and act across complex work that way. It operates across workflows workflows. Yet for most, this remains difficult to that span organizational lines: “cybersecurity translate into practice. What is often missing is a monitoring”, “document life cycle management”, credible basis for deciding where to begin, what “eligibility assessment”, “fraud detection” and to prioritize and how to move from experimentation “benefit calculation”. Harnessing agentic AI therefore to operational reality. requires a fundamentally different understanding of public administration itself. This matters because the cost of poor implementation is real.","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-4","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"b90243a4124677521baf84913266c31f063f144dab8cb07cff65f14862c9402c","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":9,"page":4,"heading":"정부의 에이전틱 AI 도입 준비 · 4쪽","text":"Governments that pursue This report develops that understanding. agentic AI without a realistic understanding It maps 70 core government workflows against of where the potential lies – and where it does two dimensions – the opportunity for agentic AI to not – risk investing in the wrong places, building add public value, and the complexity of deploying solutions that do not deliver and losing institutional it responsibly – showing where governments can confidence in the technology along the way. act with confidence, where targeted preparation Enthusiasm alone does not prevent this. What is needed, and where caution is warranted. We governments need is enthusiasm paired with offer this as a shared starting point (adaptable to strategic clarity: a grounded view of which local conditions), not a final answer. Agentic AI will government operations are genuinely suited for reshape how governments operate. Whether that agentic AI and which require a different approach. transformation is guided by evidence and strategic intent, or left to chance, is a choice. This report","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-4","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"1b903703212cfd78bb102fe1cc7e6b809b409e9c7cb68b020a6e829e56aa2c82","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":10,"page":4,"heading":"정부의 에이전틱 AI 도입 준비 · 4쪽","text":"is our contribution to ensuring it is the former.\n\nMaking Agentic AI Work for Government 4","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-4","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"ca2ee95aad408803ba6f13d04168611bc2548f914668d84c380df0e4f7d5b1e2","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":11,"page":5,"heading":"정부의 에이전틱 AI 도입 준비 · 5쪽","text":"Executive summary\nA pioneering framework to help\ngovernments move agentic AI from\nexperimentation to scalable public value.\n\nPublic expectations are rising, fiscal space for scaled adoption where institutional capacity\n is tightening, and many administrations are and safeguards are in place.\n being asked to deliver more with less. Against\n this backdrop, agentic artificial intelligence (AI) Clear takeaways emerge from this analysis:\n represents a fundamental shift in capability,\n enabling systems to autonomously execute end- – Think in functions, not departments:\n to-end multi-step workflows, with the potential to Agentic AI operates in workflows that cut\n transform how governments serve citizens. across organizational boundaries.","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-5","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"d3f0cbf4c5c77a07975210a165d963cbf64bf5e786c0658135b9730fb0f1f521","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":12,"page":5,"heading":"정부의 에이전틱 AI 도입 준비 · 5쪽","text":"Realizing this opportunity requires strategic, – Balance ambition with feasibility:\n evidence-based adoption – grounded in a clear High agentic AI potential should be\n assessment of where agentic AI can deliver the weighed against implementation complexity\n greatest public value, what risks must be managed before operationalizing at scale.\n and what capabilities and safeguards need to be\n in place before deployment at scale. – Start where the best odds exist:\n Build capability and confidence with high-\n This report responds to that need by introducing readiness functions before tackling more\n the first systematic framework for assessing complex ones.\n government readiness for agentic AI and evaluating\n core global government activities (or “functions”). – Local context determines success:\n The framework is complemented by real-world Global scores are baselines. Local infrastructure,\n use cases that ground the assessment in practice, regulatory environments and cultural norms\n highlighting current agentic AI initiatives. determine what is possible.","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-5","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"7bf0211e333b5880749334839e133335080aac8cc5150a3023caff734b2f19af","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":13,"page":5,"heading":"정부의 에이전틱 AI 도입 준비 · 5쪽","text":"– Expect the topography to evolve:\nThe starting point: assessing Reassess regularly. Functions in the low-\nagentic AI readiness across readiness area today may be in the medium- or high-readiness area tomorrow.\ngovernment functions","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-5","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"605b4c426a3c830e6be34481e24be4de9035ef999b752c3f41f2b04b35f795b9","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":14,"page":5,"heading":"정부의 에이전틱 AI 도입 준비 · 5쪽","text":"By focusing on functions – recurring workflows that From insight to action cut across organizational segregation rather than isolated tasks or departments – this framework provides governments with a new basis for The report translates analysis into a practical prioritizing deployment and implementation at scale. decision-support framework for governments considering agentic AI adoption. It provides Each function is assessed across two dimensions a structured approach to identifying relevant (agentic AI potential and implementation complexity) functions and prioritizing opportunities based and mapped into three readiness areas: on potential public value and implementation barriers, helping decision-makers focus on High-readiness area: suitable for early where agentic AI is most likely to make an deployment, with robust safeguards impact. These insights are intended to inform subsequent choices on governance design, Medium-readiness area: suitable for phased risk management, piloting and scaling. Local","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-5","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"81cbe75bce74e32f97b9545ed842a4ebf24750317d16622097ef7ff64ced50f6","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":15,"page":5,"heading":"정부의 에이전틱 AI 도입 준비 · 5쪽","text":"implementation requiring additional enabling adaptation is essential: each jurisdiction must conditions, capability building or analysis tailor the framework to its organizational priorities, digital maturity and regulatory context. Low-readiness area: suitable for monitoring, iterative testing and longer-term preparation This framework is a starting point – a tool to inform strategic choices. It is designed to help The review of 70 core government functions policy-makers and industry move together indicates that 50% combine significant agentic – establishing where to begin, how to build AI potential with manageable implementation capability and ultimately how to convert agentic complexity, pointing to a substantial opportunity AI opportunities into measurable public value.\n\nMaking Agentic AI Work for Government 5","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-5","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"92a81514aedc7354960112325588a2b2da25621a71a0f7b9d00f105284063264","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":16,"page":6,"heading":"정부의 에이전틱 AI 도입 준비 · 6쪽","text":"1 The agentic opportunity\n Agentic artificial intelligence can transform\n public institutions by orchestrating end-to-end\n workflows, shifting government operations from\n task automation to outcome delivery.\n\nMaking Agentic AI Work for Government 6","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-6","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=6","page":6},"trust":"source-data-not-instructions"},{"id":"0b35b8cae885e1c65453f29cf251b601912d4003909d8e4a4d8a264851868bcd","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":17,"page":7,"heading":"정부의 에이전틱 AI 도입 준비 · 7쪽","text":"Government technology has become a foundational Unlike earlier AI applications that focused on\n driver of competitiveness, institutional capacity narrow tasks such as classification, prediction\n and public trust. The Global Public Impact of or pattern recognition, agentic systems can\n GovTech: A $9.8 Trillion Opportunity estimates coordinate multi-step processes, integrate\n a $9.8 trillion opportunity from public-sector information across multiple sources and adapt\n digital transformation by 2034.1 Converting that their actions based on context and evolving\n opportunity into sustained impact, however, conditions. These capabilities are well aligned\n depends on how governments design and with the structure of many government\n deploy the next generation of capabilities. workflows, which typically involve sequential\n decisions, multiple stakeholders and the\n Agentic artificial intelligence (AI) is one of the application of rules and judgement over time.\n most important levers for realizing this potential.\n\n1.1 From process digitization\n to outcome orchestration","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-7","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"},{"id":"c525065233bb0ad9d0bc581a5ab7dff809d84fc4e29d0baa6e66419cd29332f6","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":18,"page":7,"heading":"정부의 에이전틱 AI 도입 준비 · 7쪽","text":"Agentic Many government digital transformation efforts to orchestrate interdependent steps can\nsystems can have focused on digitizing existing processes: enhance performance and consistency\ndeliver meaningful moving paper applications online, digitizing without a full re‑engineering effort.\nefficiency gains records and automating data entry. These\n efforts have delivered real value, but they have – Agentic reinvention: Agentic AI enableswhile maintaining\n largely preserved the underlying logic of public a structural rethink of legacy processes by\nthe accountability, administration – the medium changes, while the shifting the focus from task execution to\nquality and trust process remains fundamentally the same. Agentic outcome orchestration, opening the door to\non which public AI marks a meaningful shift because it does more new service models and operating patterns\nservices depend. than analyse data: it can plan, decide and act that were not possible under traditional,\n across multiple steps of a workflow (see Box 1). fragmented system architectures.","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-7","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"},{"id":"8a7a0dad3904514d760586d26a485dbdaf152ca4ab303b591de09f7ec63f4a71","documentId":"0fbedaadb69d81dc2a6124f4","documentSha256":"0fbedaadb69d81dc2a6124f4e6cb7422b3c233a3693f49a0b4f77f7af7c40e83","extractionVersion":"9f49d6b60cc4f1ba0cec496243cdcdc0d532500e18a3e09caf511c2eb4b0d3e6","ordinal":19,"page":7,"heading":"정부의 에이전틱 AI 도입 준비 · 7쪽","text":"Until now, public‑sector digitization has operated Crucially, agentic AI is best understood as a model of under a familiar constraint: digitizing a flawed augmentation rather than replacement. By handling process largely preserved its flaws. Agentic AI can routine coordination, accelerating case processing help alleviate this constraint by reasoning across and surfacing relevant information, these systems information and executing goal-directed actions. allow public servants to focus on meaningful work While processes must still be redesigned to unlock that requires human judgement: complex cases, its full potential, improvements to existing workflows policy interpretation and direct engagement with can deliver meaningful gains in the near term. citizens. When deployed with appropriate oversight, governance and safeguards, agentic systems can – Agentic workflow optimization: Instead of deliver meaningful efficiency gains while maintaining requiring a complete process redesign before the accountability, quality and trust on which","citation":{"title":"정부의 에이전틱 AI 도입 준비","publisher":"World Economic Forum","url":"/library/0fbedaadb69d81dc2a6124f4#page-7","pdfUrl":"/library/0fbedaadb69d81dc2a6124f4/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"}]}