2026年人工智能（AI）在学习与发展领域报告

AI in Learning &Development Report 2026 Executive SummaryExecutiveSummary 0101 Our L&D ExpertsOurL&DExperts 0303 Survey MethodologySurvey Methodology 0404 Moving Past ExperimentationMoving Past Experimentation 0606 How L&D Is Using AI: Today & Tomorrow09How L&D Is Using Al:Today & Tomorrow09 Current Value vs. Future Value11Current Valuevs.Future Value11 Budgets, Readiness & Blockers13Budgets,Readiness&Blockers13 The Future Ecosystem & Agentic AI16TheFutureEcosystem&AgenticAl16 What’s Next for L&D Teams?19What'sNextforL&DTeams?19 Insights from Dr Philippa HardmanInsightsfromDrPhilippa Hardman 2121 ExecutiveSummaryExecutiveSummary AI has moved beyond experimentationand into everyday L&D workAlhasmovedbeyondexperimentationand intoeverydayL&Dwork 87% of respondents are already using AI, and only 2% have noadoption plans. Most are past experimentation, with 36% usingAI in defined workflows and 9% beginning to scale it acrosstheir organization.87% of respondents are already using Al, and only 2% have noadoption plans. Most are past experimentation, with 36% usingAl in defined workflows and 9% beginning to scale it acrosstheir organization. AI’s strongest value today is speedAl's strongest value today is speed Today L&D teams use AI mainly for voice generation (63%),content and quiz drafting (60%), video creation (52%) andtranslation (38%) in the design and development stages. Thetop benefits are faster production (84%) and better learnerexperience (66%).Today L&D teams use Al mainly for voice generation (63%),content and quiz drafting (60%), video creation (52%) andtranslation (38%) in the design and development stages. Thetop benefits are faster production (84%) and better learnerexperience (66%). “AI has crossed a threshold in L&D. It has movedfrom experimental tool to everyday practice — and,for a growing minority, to something closer tooperational infrastructure.” "Al has crossed a threshold in L&D. It has movedfrom experimental tool to everyday practice and,for a growing minority, to something closer tooperational infrastructure." -DR. PhiliPpa Hardman-DR, PHILIPPA HARDMAN The next phase is about learner impactThe nextphase isaboutlearnerimpact L&D teams expect the biggest future gains in morepersonalized learning (72%), wider internal reach (65%) andimproved learner engagement (56%).Planned adoption is risingfor assessments and simulations (36%), adaptive pathways(33%), skills mapping (32%) and AI tutors (29%). And value isshifting from time saved (88%) to clearer business impact(55%) and easier global localization (54%).L&D teams expect the biggest future gains in morepersonalized learning (72%), wider internal reach (65%) andimproved learner engagement (56%). Planned adoption is risingfor assessments and simulations (36%), adaptive pathwaysS n %)e (%e) de (%)shifting from time saved (88%) to clearer business impact(55%) and easier global localization (54%). The learning ecosystem is becomingmore distributedThe learning ecosystem is becomingmore distributed Only 47% think the Learning ManagementSystem (LMS) willremain the backbone of their ecosystem.Expectations forwhere AI will live split across embedded features (19%),productivity tools (17%), standalone systems (17%) and cross-system agentic layers (19%). 27% remain unsure.Only 47% think the Learning Management System (LMS) willremain the backbone of their ecosystem. Expectations forwhere Al wilive split across embedded features (19%).productivity tools (17%), standalone systems (17%) and cro8s-system agentic layers (19%). 27% remain unsure. Agentic AI (AI that can autonomously take actions)interest is high, but teams are cautiousAgentic Al (Althatcanautonomouslytakeactions)interestishigh,butteamsarecautious Most are excited (27%), cautious (39%) or say they need to learn more (29%). Explorationfocuses on AI tutors (49%), coaching and mentoring (43%), personalized guidance (43%) andadmin automation (38%).Most are excited (27%), cautious (39%) or say they need to learn more (29%). Explorationfocuses on Al tutors (49%), coaching and mentoring (43%), personalized guidance (43%) andadmin automation (38%). Budgets remain low and fragmentedBudgets remain low and fragmented 39% spend 5% or less of their L&D budget on AI, and 30% don’t know their spend. Dedicatedfunding is still developing.39% spend 5% or less of their L&D budget on Al, and 30% don't know their spend. Dedicatedfunding is still developing. Readiness challenges are slowing progressReadiness challengesare slowingprogress Security (58%), accuracy (52%), legal constraints (41%) and integration challenges (36%)remain major obstacles. Although 74% say their culture encourages experimentation, only45% feel IT is enabling AI adoption.Security (58%), accuracy (52%), legal constraints (41%) and integration challenges (36%)45% feel IT is enabling Al adoption. We finally have the tools to create dynamic learning content at a speed andscale we’ve never seen before. The opp