{"schemaVersion":"1.0","document":{"id":"f2bac170b8f99defb7094ef0","title":"사이버보안을 위한 AI","originalTitle":"Empowering Defenders AI for Cybersecurity","category":"digital","categoryLabel":"디지털·보안","publisher":"World Economic Forum","language":"en","publicationYear":2026,"publicationMonth":5,"publishedOn":"2026-05-04","datePrecision":"day","dateSource":"https://www.weforum.org/publications/empowering-defenders-ai-for-cybersecurity/","coverUrl":"/assets/covers/library-f2bac170b8f99defb7094ef0.jpg","tags":["사이버보안","보안","AI 방어","cybersecurity","defenders"],"pages":34,"bytes":6030125,"sha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","searchable":true,"url":"/library/f2bac170b8f99defb7094ef0","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf","chunksUrl":"/api/library/f2bac170b8f99defb7094ef0/chunks","extraction":{"totalPages":34,"pagesWithSubstantialText":34,"sparsePages":[],"method":"pdf-text","ocrPerformed":false},"relatedPublications":[{"series":"커스텀 브리프","issue":"2호","url":"/archive/custom-02","title":"AI가 먼저 뚫어본다, 미토스 쇼크 그 이후"},{"series":"주간 WEF","issue":"2호","url":"/archive/davos-02","title":"시험 중인 학생이 선생님 서랍을 열었다 — AI 에이전트가 자기 시험의 정답지를 훔친 사건, 그리고 네 개의 공백"},{"series":"주간 WEF","issue":"8호","url":"/archive/davos-08","title":"고른 사람이 없는 칸 — 자율성과 권한은 설계 선택이다"}],"metadataNote":"한국어 제목·분류·검색어는 원문 탐색을 위한 편집 정보입니다."},"extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","total":123,"offset":0,"nextOffset":20,"chunks":[{"id":"4e768d5016ff700ec2fb0325d17c633f5b0c4d68950c630ed20e08a6e83ea323","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":0,"page":1,"heading":"사이버보안을 위한 AI · 1쪽","text":"In collaboration with KPMG\n\nEmpowering Defenders:\nAI for Cybersecurity\n\nW H I T E P A P E R\nM A Y 2 0 2 6","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-1","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=1","page":1},"trust":"source-data-not-instructions"},{"id":"29821805f4e97300c46e714b7a7de33838f1c55936c7643a51389e67a8629265","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":1,"page":2,"heading":"사이버보안을 위한 AI · 2쪽","text":"Images: Getty Images\n\nContents\n\nForeword 3\n\nExecutive summary 4\n\nIntroduction 5\n\n1 Present-day uses of AI in cybersecurity 7\n\n1.1 AI-driven cyber governance 8\n\n1.2 AI-enabled risk identification 9\n\n1.3 AI-augmented cyber protection 11\n\n1.4 AI-powered threat detection 14\n\n1.5 AI-orchestrated incident response 17\n\n1.6 AI-supported incident recovery 20\n\n2 Important considerations for AI adoption in cybersecurity 21\n\n2.1 How does the adoption of AI for cybersecurity align with 22\n and accelerate strategic priorities?\n\n2.2 Is the organization ready to deploy AI in cybersecurity 22\n effectively?\n\n2.3 How can the organization validate AI solutions before 24\n full deployment?\n\n2.4 How best to scale and maintain AI solutions while ensuring 25\n continuous optimization?\n\n3 The evolving landscape of AI in cybersecurity 26\n\n3.1 The opportunity of agentic AI 26\n\n3.2 Unpacking the spectrum of AI autonomy 26\n\n3.2 Agentic AI risks and guardrails 27\n\nConclusion 28\n\nContributors 29\n\nEndnotes 33","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-2","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=2","page":2},"trust":"source-data-not-instructions"},{"id":"aaf15760c8aacb355745d2b76dc3d4e6f3a78201bed5d0bb9eb4eec1537c9f9f","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":2,"page":2,"heading":"사이버보안을 위한 AI · 2쪽","text":"Disclaimer\nThis document is published by the World Economic\nForum as a contribution to a project, insight area\nor interaction. The findings, interpretations and\nconclusions expressed herein are a result of a\ncollaborative process facilitated and endorsed by\nthe World Economic Forum but whose results do\nnot necessarily represent the views of the World\nEconomic Forum, nor the entirety of its Members,\nPartners or other stakeholders.\n© 2026 World Economic Forum. All rights reserved.\nNo part of this publication may be reproduced or\ntransmitted in any form or by any means, including\nphotocopying and recording, or by any information\nstorage and retrieval system.\n\nEmpowering Defenders: AI for Cybersecurity 2","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-2","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=2","page":2},"trust":"source-data-not-instructions"},{"id":"32594f72cbc12b09883fbf71e9bd46b5e0fbec0ef5e9b7d03fc2b4eca0bc1ec8","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":3,"page":3,"heading":"사이버보안을 위한 AI · 3쪽","text":"May 2026 Empowering Defenders:\n AI for Cybersecurity\n\nForeword\n\nAkshay Joshi\n Head of the Centre for Laurent Gobbi\n Cybersecurity; Member of Partner, Global Head of\n the Executive Committee, Cyber & Tech Risk, KPMG\n World Economic Forum","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-3","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"7b52b3c0bb35f9e75e92bff79b10b1d76445e8db8833569f31e5cbccd1d73c8f","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":4,"page":3,"heading":"사이버보안을 위한 AI · 3쪽","text":"With the advent of the Intelligent Age, artificial Success will depend on key foundational elements, intelligence (AI) and cybersecurity have become including strong executive support, high- inextricably linked. In 2025, the World Economic quality data, a skilled workforce and integrated Forum published Artificial Intelligence and infrastructure. As automation becomes increasingly Cybersecurity: Balancing Risks and Rewards, prevalent, preserving human judgement and which highlighted the inherent cybersecurity expertise remains essential to mitigate potential risks associated with AI adoption and provided systemic fragility. organizations with a roadmap to mitigate them. In this paper, the focus shifts to the transformative potential Developed by the World Economic Forum’s Cyber of AI in strengthening cybersecurity defences. Frontiers initiative in collaboration with KPMG, this paper brings together contributions from leading This white paper examines the application of AI experts in cybersecurity and AI. Through a series","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-3","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"ffb4c457774881de5451cde14fab449097531de50298d93952f7f1fda0976ea6","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":5,"page":3,"heading":"사이버보안을 위한 AI · 3쪽","text":"technologies across the cybersecurity life cycle, of workshops, these leaders shared insights into drawing on real-world examples from World how AI can be harnessed effectively to enhance Economic Forum partner organizations. It outlines cyber resilience. As AI becomes further embedded the strategic considerations and practical steps that within cybersecurity operations, strategic alignment executives and chief information security officers will be critical to staying ahead of increasingly (CISOs) must undertake to enable the effective sophisticated cyber threats and to supporting adoption and deployment of AI in cybersecurity. sustainable growth in the digital economy.\n\nEmpowering Defenders: AI for Cybersecurity 3","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-3","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=3","page":3},"trust":"source-data-not-instructions"},{"id":"7f279e2ec4daa124a8d936d62eb6a77cee2882305810c86b1ba38ba50ac93ee9","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":6,"page":4,"heading":"사이버보안을 위한 AI · 4쪽","text":"Executive summary\nWhile AI can enhance cyber\ndefences, its effectiveness relies on\nhaving a clear strategy grounded in\nhuman oversight and judgement.","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-4","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"5c167e1a09beb38fb7265b2b966290477055c203317e31ea8d27adf39fbb3ee8","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":7,"page":4,"heading":"사이버보안을 위한 AI · 4쪽","text":"Artificial intelligence (AI) is reshaping the Executive and cyber leaders embarking on the AI cybersecurity landscape. Attackers are increasingly adoption journey for cyber defence should therefore: using AI to increase the speed, scale and sophistication of threats. To address these evolving – Align the adoption of AI in cybersecurity with risks, cybersecurity must keep pace with the organizational strategic priorities growing speed and sophistication of modern attacks. AI-driven tools are becoming central across – Establish organizational readiness across the cybersecurity life cycle, from detecting and processes, data, infrastructure, skills and preventing incidents to response and recovery, governance before deploying AI in cybersecurity enabling organizations to secure their digital assets and sensitive information more effectively. – Validate AI solutions through structured pilots prior to full deployment This paper focuses on the use of AI by defenders, thereby providing actionable guidance to – Scale and monitor the performance of AI in","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-4","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"fac6ae48d2d57e967ee90a81b813e769d049a3e9fa30c09b1aee96a63b7ec4c6","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":8,"page":4,"heading":"사이버보안을 위한 AI · 4쪽","text":"organizations on how to deploy AI to enhance cybersecurity and optimize as needed defensive capabilities. With the right approach in place, organizations can Real-world applications of AI in cybersecurity, harness AI technologies to strengthen defences illustrated through case studies submitted by today, while building the agility to keep pace with World Economic Forum partners, demonstrate, evolving cyberthreats. for example, the use of AI to improve vulnerability detection, curate threat intelligence data and enhance defences against phishing and other attack vectors. This reflects a shift to operational adoption of AI solutions delivering measurable improvements in security performance.\n\nEmpowering Defenders: AI for Cybersecurity 4","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-4","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=4","page":4},"trust":"source-data-not-instructions"},{"id":"cbf8b9a5bbe299914b5d1d4219b67bd5db80416498a6f68b6a471b0e3c32ef06","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":9,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"Introduction\nThe deployment of AI in cybersecurity\nis accelerating, although the pace and\ndepth of adoption vary considerably.","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"a1c73b4525d8e99792a81f80e5cb6282a069ae1b1ca5e14e9a0373c84177c801","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":10,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"AI is no longer a peripheral innovation; it is redefining breach costs by $1.9 million.3 More broadly, AI cybersecurity by enabling advanced defence helps address the structural challenges faced by capabilities, creating new requirements to secure cybersecurity, including the rising volume and AI-driven systems and amplifying the tools available sophistication of attacks, persistent talent shortages to adversaries. In fact, its strategic importance is and increasing system complexity. reflected in the findings of the Global Cybersecurity Outlook 2026, which identifies AI as the most Yet progress on the defensive side is unfolding significant driver of change in cybersecurity, in parallel with a rapidly evolving threat according to 94% of survey respondents. environment, where adversaries are increasingly operating at machine speed, using AI to conduct While 77% of organizations report using AI reconnaissance of targets and vulnerabilities, in cybersecurity, deployment is closely tied generate malware, exploit code, evade detection","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"93ad4a5fcf34e30ccbb0201ee78e1a7e92e89463f8e8904764a75b82f0a455a8","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":11,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"to organizational size and resources.1 Larger and launch attacks at scale. What once required companies, supported by greater technical maturity weeks of effort can now be executed in minutes, and investment capacity, report higher adoption lowering technical barriers and dramatically rates, whereas smaller entities, governments and expanding both the volume and impact of non-governmental organizations (NGOs) tend to lag cyberattacks.4 behind due to financial constraints, skills availability and data maturity. Against this backdrop, AI has become a core enabler of modern cybersecurity. Its value lies When deployed effectively, AI delivers measurable not just in automation but in augmenting human operational and financial benefits: 88% of security cognition and expertise, accelerating detection and teams report time saving and greater opportunity decision-making and strengthening organizational for proactive defence.2 Organizations using AI preparedness across technical, operational and extensively in security shortened breach times governance dimensions.","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"9410649876ac879ca257ff4709e150c57aaa79f9747d4b8dcc6be2251b49f44b","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":12,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"by approximately 80 days and reduced average\n\nWhy is AI for cybersecurity a strategic necessity?","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"a02f9c930f3fa476aa703120f2c2a6d25ae5afdf31c00fa41857e1af71ea87cc","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":13,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"The rising complexity, volume and speed of Complexity and scale: Modern digital ecosystems cyberthreats is outpacing traditional defences, are highly interconnected, creating large attack requiring new approaches to protect interconnected surfaces and hidden vulnerabilities. Manual analysis systems. AI has emerged as both a technical and is no longer sufficient. AI can operationalize and strategic capability to address these challenges, correlate data in real time, identifying configuration enabling organizations to strengthen defensive weaknesses and vulnerabilities at a scale and operations and meet regulatory demands. Several speed beyond human capability. factors underscore why AI has become essential for modern cyber defence: Operational burden: Security teams are often overwhelmed by alert overload and repetitive tasks, Defensive asymmetry: Attackers need only find one with 76% of professionals reporting exhaustion entry point while defenders must defend everything. in 2025.6 AI can automate routine processes and With AI, attackers can identify vulnerabilities faster.","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"36be32acbaefa5e6cb477aa67ca975b30005e82ccc7fce6842f14d5f55960c6f","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":14,"page":5,"heading":"사이버보안을 위한 AI · 5쪽","text":"prioritize and filter alerts, freeing teams to focus on Defenders, however, can use AI to analyse and high-value, proactive cybersecurity activities. prioritize risks using internal proprietary data for contextual precision that attackers lack, thereby regaining strategic advantage.5\n\nEmpowering Defenders: AI for Cybersecurity 5","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-5","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=5","page":5},"trust":"source-data-not-instructions"},{"id":"086b15d8674a6c38f96e4bd1f67bf14d8ee0f406e0b1cfc604609a5590a397f5","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":15,"page":6,"heading":"사이버보안을 위한 AI · 6쪽","text":"Resource constraints: Expanding digital footprints reporting and mitigation of security incidents. As\noften outpace cybersecurity budgets and staffing. requirements expand in scope and complexity, AI\nIn 2025, 53% of teams reported underfunding and can act as a force multiplier, accelerating incident\n55% reported understaffing.7 AI can help augment detection and response, automating monitoring\nhuman expertise, scale security operations and documentation and maintaining regulatory\nefficiently and help address talent gaps. readiness across the organization without\n increasing operational burden.\nRegulatory and compliance pressures:\nCybersecurity regulations – whether industry- Ultimately, organizations that approach AI in\nspecific, such as the Digital Operational Resilience cybersecurity strategically, aligning governance\nAct (DORA), or cross-industry, such as the and operations, will be best positioned to counter\nNIS2 Directive – mandate the rapid detection, adversaries and mitigate misuse.\n\nGuarding against over-reliance on AI in cybersecurity","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-6","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=6","page":6},"trust":"source-data-not-instructions"},{"id":"f0991f257164a78a757fa82107a22dc414069fddbe37da1d85bd3a9f0a018982","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":16,"page":6,"heading":"사이버보안을 위한 AI · 6쪽","text":"Heavy reliance on AI can undermine cyber resilience. Excessive trust in automated\n decisions creates a false sense of security and over time erodes the expertise needed\n to intervene when systems fail.\n\nTo prevent over-reliance on AI, security teams should combine AI with human judgement,\n simulate AI failures and design fail-safes that keep security operations functional during\n AI outages.\n\nEmpowering Defenders: AI for Cybersecurity 6","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-6","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=6","page":6},"trust":"source-data-not-instructions"},{"id":"66efa5758c88abf84197d36374cc1e9fca4801f6e05522a5561f30558e8b7f3a","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":17,"page":7,"heading":"사이버보안을 위한 AI · 7쪽","text":"1 Present-day uses of\n AI in cybersecurity\n AI in cybersecurity spans multiple\n functions, supporting activities across\n the entire security life cycle – from\n identifying risks and protecting systems\n to detecting threats, responding to\n incidents and enabling recovery.","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-7","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"},{"id":"41579a4c80f3c6c27765ab5ef82880fd8b8b0406a10de865b3a1d8c63ea6ebcb","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":18,"page":7,"heading":"사이버보안을 위한 AI · 7쪽","text":"To understand the current state of AI in Many case studies span multiple functions, reflecting cybersecurity, this section draws on case studies the interconnected nature of cybersecurity activities. submitted by World Economic Forum partners While not exhaustive, the concentration of case from various sectors and geographies, highlighting studies in the identify and detect functions is practical applications and real-world experience. consistent with the Global Cybersecurity Outlook 2026, which finds that 52% of organizations use AI While the case studies are mapped against the six for phishing detection, 46% for intrusion and anomaly functions of the NIST Cybersecurity Framework 2.0 detection and 40% for user behaviour analytics. (see Figure 1) – govern, identify, protect, detect, respond and recover8 – it is important to note that Additional insights come from workshops the framework is not applied prescriptively but conducted under the World Economic Forum’s rather serves to structure and categorize the case Cyber Frontiers initiative, convening 105","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-7","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"},{"id":"f1a82caa4e72b5ee64926c1685140d47e1948d0c61a80189214bbf71dd91d37a","documentId":"f2bac170b8f99defb7094ef0","documentSha256":"f2bac170b8f99defb7094ef00d0bcf0362ad7453806933e79c2487ad5dbbaf7b","extractionVersion":"d23b44fb8807fdbad1da6713adaa872376b18e7e950e3f171ca771341ef13c8d","ordinal":19,"page":7,"heading":"사이버보안을 위한 AI · 7쪽","text":"studies, ensuring that they are presented within a representatives from 84 organizations across coherent and standardized context. 15 industries.\n\nF I G U R E 1 The six functions of the NIST Cybersecurity Framework9\n\nGovern Identify Protect\n Manage risk and align Map the digital environment Establish safeguards and\n security controls with legal to inventory assets and assess technical controls to prevent\n and business requirements organizational cyber risks attacks and limit impact\n\nDetect Respond Recover\n Enable continuous Restore operations while Contain and neutralize threats\n monitoring to recognize strengthening system while managing impact and\n suspicious patterns and configurations and stakeholder communication\n breaches in real time organizational resilience\n\nSource: World Economic Forum\n\nEmpowering Defenders: AI for Cybersecurity 7","citation":{"title":"사이버보안을 위한 AI","publisher":"World Economic Forum","url":"/library/f2bac170b8f99defb7094ef0#page-7","pdfUrl":"/library/f2bac170b8f99defb7094ef0/original.pdf#page=7","page":7},"trust":"source-data-not-instructions"}]}