Maybank, Pearson Execs Shift HR Strategy to AI Workflows

Maybank与Pearson高管将HR战略转向AI工作流 企业AI落地项目有70%会面临失败，除非组织重构激励机制、打破遗留数据孤岛，并围绕人机协同重新设计工作岗位。 该研讨小组探讨了随着组织从传统层级架构向人机协同转型，劳动力战略发生的根本性转变。主要发言人讨论了如何通过组建跨职能特别工作组、重塑非货币激励体系以及实现孤立数据栈的普及与共享，从而超越浅层的AI试点项目。他们强调，要扭转AI项目约70%的失败率，需要高管发挥强有力的示范带头作用（如举行C-suite高管技术集训研讨），并将AI工具作为切实的“止痛药”直接嵌入日常工作中，以此提升人类的判断力与顾问价值。 传统的管理层战略三角（CEO、HR、财务）正在演变为菱形架构，机器智能已被纳为核心运营支柱。 (0:04) 企业领导者现在期望HR能够先行推动由AI赋能的生产力提升并重新设计核心工作流，而非在执行后才让员工去适应传统遗留系统。 (4:50) AI的普及正在改变人才招聘标准：对候选人的评估侧重于AI适应能力和非传统的影响力佐证，同时HR团队正越来越多地引入工程师和数据分析师来搭建AI基础设施。 (4:50) 在整个员工生命周期中保持人类价值，需要招聘人员对自动化系统保持监督，以持续的能力建设取代偶发性的阶段培训，并践行基于价值观的招聘。 (13:54) 企业衡量AI投资回报率的途径主要包括运营效率（在人员编制不呈线性增长的前提下实现业务规模化扩展）和绩效影响（将低价值任务自动化，让员工专注于高价值工作）。 (20:00)

In an Ai4 panel moderated by Dr. Jana Brown, workforce and technology leaders argue that integrating artificial intelligence transforms the traditional executive triangle of CEO, HR, and finance into a four-pillar diamond that includes machine intelligence. Capturing AI's value requires organizations to look beyond software deployment to fundamentally redesign workflows, incentives, and leadership culture. Redesigning Work and the Strategic HR Mandate (4:50) C-suite leaders no longer view human resources merely as an operational support function; they expect HR to drive AI-enabled productivity from the outset. Esin Cooper of Pearson notes that rather than adapting workers to established systems midway through execution, executives now ask HR to help redesign work itself—determining which tasks should remain human-led, which should be augmented, and which should be automated. To support this shift, Giovanni Luperti of Humaa observes that traditional assumptions about management layers and headcount are becoming obsolete as engineers and data analysts join HR teams to co-create infrastructure alongside CIOs and CTOs. At the same time, workforce strategy must protect high-value human contributions. Drawing on an 8-to-10-year digitalization of 40,000 Maybank employees across Southeast Asia, Shameem Farouk argues that workforce strategy cannot be separated from business strategy. In sectors like banking, automating routine tasks is meant to restore meaning and motivation so employees can focus on complex client advisory services that rely on tacit knowledge current AI cannot replicate. Balancing Automation and Judgment in Talent Strategy (13:54) AI is reshaping how organizations acquire and develop talent while raising the premium on human judgment. Rosie Buller of JPMorgan Chase explains that while AI helps non-technical recruiters translate complex engineering requirements into plain language, over-reliance on automated applicant tracking systems (ATS) is risky. An ATS may reject a qualified candidate for missing specific résumé keywords, whereas a human recruiter checking Google Scholar can identify relevant PhD research or patents. Conversely, hiring managers now evaluate candidates beyond two-page résumés, looking for adaptability and how effectively they combine human judgment with AI outputs. Once employees are hired, episodic training is giving way to continuous capability building and flexible role design. Pearson introduced "Pearson Power Skills" to cultivate adaptability, learning agility, and AI proficiency across its career architecture. Farouk adds that organizational resilience depends on hiring for values and critical thinking rather than rigid, task-based job descriptions—citing a Maybank team that built and launched an e-commerce platform for small and medium enterprises in three weeks during the COVID-19 pandemic without formal job descriptions. Targeting ROI and Avoiding Implementation Traps (30:31) According to Luperti, enterprises measure AI return on investment through two primary lenses: operational efficiency—scaling output without linearly increasing headcount—and performance impact, which frees top performers from tedious work. Yet the panel notes that roughly 70% of AI initiatives fail to achieve their objectives. Projects typically stall when companies spread efforts across superficial, checkbox pilots or treat AI purely as a technology rollout without updating governance, delegation of authority, and team structures. High-performing organizations instead focus deeply on high-friction operational areas, such as automating end-to-end employee onboarding, or embed AI into daily routines as an immediate "painkiller." For example, Pearson deployed an AI agent named "Bar Up" to help managers and employees draft Objectives and Key Results (OKRs) and performance reviews, reducing administrative overhead so leaders can focus on coaching. Overcoming Internal Friction Through Coalitions and Role-Modeling (23:56) Transforming legacy organizations often triggers resistance from both business units absorbed in daily operations and internal IT teams, which naturally resist high-risk experimentation because they are evaluated on 100% system uptime and reliability. Regulated institutions also face strict compliance hurdles when attempting to break down functional data silos to feed enterprise machine learning models. To overcome this inertia, speakers advocate forming agile, cross-functional task forces across HR, tech, and business units, backed by career promotion incentives for collaborative champions rather than one-off monetary bonuses. Sustaining adoption also requires visible executive role-modeling. Pearson mandates high AI proficiency standards for senior executives tied to their job tiers. Similarly, Farouk recalls that during Maybank's early machine learning adoption around 2017–2018, the CEO, C-suite, and board completed a 2.5-day retreat to learn Python coding and algorithm design firsthand—a display of vulnerability that shifted the corporate culture from a "know-it-all" to a "learn-it-all" mindset. Because generative AI lacks decades of validated playbooks, leaders must navigate ambiguity and learn across hierarchies, including from junior engineers and interns who work closest to emerging tools. By combining learning agility, ethical governance, and a cross-functional executive coalition uniting the CEO, finance, technology, and business heads, organizations can pursue what Indiana University research terms "enlightened intelligence"—using machine capabilities to elevate human purpose and shared prosperity.

Ai4专家研讨会指出，人工智能正促使企业传统管理三角向包含机器智能的菱形架构转变，人力资源与业务战略必须深度融合以重构工作流并防范高达70%的项目执行失败率。 人机协同重塑组织架构与HR职能 (0:04) 研讨会主持人Jana Brown博士指出，企业原先由CEO、人力资源与财务部门构成的传统管理三角，正在因人机协作融合演变为包含机器智能核心支柱的菱形架构。Maybank前高管Shameem Farouk与Pearson组织发展负责人Esin Cooper强调，劳动力战略已不能脱离业务战略独立存在。高管团队对HR的诉求已从日常行政与系统执行，转向在项目初期主导重新设计工作流程，清晰划分完全自动化、AI辅助增强以及必须由人类主导的业务边界。以银行高净值客户咨询为例，复杂的顾问服务高度依赖人类专长与隐性知识，这是现有AI无法替代的。与此同时，Humaa负责人Giovanni Luperti与JPMorgan Chase招聘副总裁Rosie Buller提到，工程开发与数据分析等技术人员正加速进入HR团队，与CIO及CTO共同搭建底层基础设施，招聘侧也开始将候选人的AI适应性与实际业务产出纳入考核标准。 重塑员工生命周期并平衡AI工具与人类判断 (13:54) 在员工生命周期的具体实践中，专家警告过度依赖自动化可能带来筛选盲区。Rosie Buller指出，若完全依赖申请人追踪系统（ATS），系统容易因简历缺乏特定关键词而剔除优秀人才，而人工招聘人员则可通过检索Google Scholar上的博士研究成果或专利发现实际能力。此外，AI工具也能反向辅助非技术招聘人员拆解专业技术栈，使其能够更准确地评估工程师候选人。在组织能力建设方面，Pearson改变了传统的阶段性培训，推出聚焦适应能力、持续学习与AI素养的“Pearson核心技能”（Pearson Power Skills），并将AI熟练度标准纳入高级管理层的层级考核中。为减少日常行政负担，Pearson还部署了名为“Bar Up”的AI智能体，协助员工与主管撰写OKR、绩效考评与自我总结，使管理者能将精力集中于高价值的绩效辅导。Shameem Farouk则通过Maybank在疫情期间仅耗时三周即上线中小企业电商平台的案例指出，企业应重视价值观与高阶思考等持久技能，而非受限于刻板的岗位描述。 AI投资回报衡量与破解70%失败率的关键 (20:00) 针对AI投资的高昂成本，Giovanni Luperti列出衡量ROI的两大核心指标：一是在不按比例增加员工人数的前提下实现业务运营规模化扩展，二是将高绩效员工从繁琐的低价值任务中解脱出来以聚焦高影响力业务。与会专家指出，约70%的AI项目因被狭隘视作单纯的软件部署而未能达到预期目标。高性能企业避免浅尝辄止的打勾式试点，而是选取高摩擦力的具体业务场景（如全流程员工入职自动化）进行端到端重构。数字化转型常常遭遇技术部门与业务部门的双重阻力：IT团队出于系统可用性与100%正常运行时间的考核要求，天然规避高风险实验；业务人员则因日常运营压力缺乏变革动力。为此，企业需组建涵盖HR、技术与业务负责人的敏捷跨职能工作组，打通各部门的数据孤岛以支持机器学习模型运算，并通过职业晋升与公开认可等非物质激励，表彰具备跨学科协作与技术应用能力的数字化骨干。 高管躬身示范与应对不确定性挑战 (33:54) 推动全员转型的先决条件在于高管的亲自示范与敢于展现脆弱性。如果领导层仅口头推崇AI却不亲自尝试，全员应用便会停滞。Shameem Farouk分享了Maybank在2017至2018年期间的经验：由CEO及董事会成员带头参与为期两天半的封闭式集训，亲自学习Python编程、算法逻辑与AI底层机制。这种自上而下公开学习过程的做法，有效破除了冒名顶替综合征，推动企业文化从“无所不知”向“无所不学”转变。面对生成式AI缺乏成熟案例与标准化操作手册的现状，管理层必须学会适应不确定性并容忍计算过的风险，打破层级壁垒直接向贴近前沿应用的初级工程师甚至实习生学习。小组引用印第安纳大学提出的“启蒙智能”（enlightened intelligence）理念，呼吁建立健全数据治理与道德框架，确保技术在提升组织学习敏捷度的同时，强化人类价值与伦理导向。 研讨会总结指出，面对AI带来的快速技术变革，HR绝不能孤立运作，必须与CEO、财务、技术及各业务线高管结成紧密联盟，将劳动力战略与企业财务及业务目标深度对齐。然而，关于多部门数据共享中的合规与风险边界如何平衡，以及AI智能体在多大程度上能够完全替代基层管理职能，仍是有待各企业在实践中持续验证的开放议题。

## Background 本内容由频道 Ai4 于 2026 年 8 月 22 日在 YouTube 平台发布。视频记录了一场关于人工智能时代劳动力战略重塑的专家圆桌论坛，汇聚了来自培生集团（Pearson）、马来亚银行（Maybank）、eHealth 以及 AI 创业公司 Humaa 的高管与 HR 领袖，重点探讨组织如何平衡技术创新与人性关怀、利用 AI 弥补技能差距并构建繁荣的人机协同团队。 ## TL;DR 圆桌论坛深入剖析了 AI 技术如何颠覆传统的企业治理与人才管理范式，推动管理结构从“CEO-HR-财务”的传统铁三角向纳入机器能力的“四角菱形”跃迁。嘉宾们强调劳动力战略即业务战略，企业必须摆脱“打卡式”的技术试点，通过跨职能敏捷行动组、打破数据孤岛与重构非物质激励来推动组织变革。同时，招聘与人才发展需坚持“人在回路中”（Human-in-the-loop），从偶发式培训转向以“学习敏捷度”和“AI 适应力”为核心的持续能力建设。面对不确定性，领导者应放下“无所不知”的权威姿态，以终身学习和展示脆弱来激发全员创新，将技术红利转化为推动人类向高价值链跃升的动力。 ## Atmosphere & style 整场对话气氛务实、睿智且充满思辨张力，既有大型跨国银行与教育巨头的落地实操复盘，也有前沿 AI 创业者的敏锐洞察。主持人以幽默轻松的节奏引导讨论，嘉宾们毫无保留地分享了转型中遭遇的技术阻力与组织阵痛，消解了空洞的技术崇拜。在严肃探讨 ROI 衡量、数据治理与组织重构的同时，对话始终贯穿着对“人性温度”与使命感的坚守，传递出一种在剧烈技术变革中清醒前行、兼具理性与同理心的领导力力量。 ## Key takeaways - [00:04](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=4s) 传统由 CEO、HR 与财务构成的管理“铁三角”正在演变为纳入机器与 AI 维度的“组织菱形”，企业核心挑战在于如何实现人机深度协同并重塑劳动力战略。 - [01:31](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=91s) 劳动力战略的本质发生根本转变：从过去的“在正确时间把合适的人放在正确岗位”转变为“围绕工作本身，将人类能力、AI 智能与学习能力进行最优组合与协同设计”。 - [04:50](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=290s) “劳动力战略即业务战略”，高价值业务深度依赖顶尖人才的隐性知识与同理心，这是当前 AI 无法复刻的人类核心壁垒。 - [08:11](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=491s) 组织发展（OD）重心从“帮助员工适应既定系统”转向“重构人机系统”，在需要高阶人类判断力的核心链条坚持 Human-in-the-Loop，辅助环节最大化释放 AI 生产力。 - [09:55](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=595s) 招聘与人才评估不再只看静态技能，而是重点考察候选人的“AI 适应力”、批判性思维以及将技术转化为实际业务成果的能力。 - [09:55](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=595s) 传统 HR 模式正在终结：CHRO 与 CTO/CIO 深度协同构建全流程 AI 基础设施，工程与数据人才直接入编 HR 团队，推动组织架构向敏捷高效转型。 - [15:48](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=948s) 人才培养从偶发、断点式培训转向持续性能力建设，核心在于打造以“学会如何学习、动态适应力、AI 熟练应用度”为支柱的动力技能（Power Skills）。 - [19:57](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1197s) 衡量 AI 投资回报率（ROI）聚焦于两大维度：在不大幅扩张人员编制的前提下实现运营效率倍增，以及剥离冗余杂务以释放核心人才的卓越绩效。 - [21:53](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1313s) 约 70% 的 AI 项目折戟在于仅将其视作“打卡式技术部署”；成功企业通常直击高价值具体痛点（如全流程入职自动化）并深度推进组织与变革管理。 - [25:56](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1556s) 单纯的员工技能再培训（Reskilling）难以独自奏效，必须依托激励机制重构、底层技术与数据民主化共享，以及跨职能组织架构设计的协同支撑。 - [35:25](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2125s) 领导力心态需从“无所不知”（Know-it-all）转向“终身学习”（Learn-it-all），高管通过展现脆弱与亲自实践学习，能有效打破全员对新技术的畏难情绪。 - [38:38](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2318s) 在生成式 AI 缺乏现成经验的背景下，领导者必须适应模糊性与不确定性，并打破层级孤岛、向下向年轻一代员工逆向学习前沿工具应用。 - [41:40](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2500s) 未来繁荣组织的核心竞争壁垒在于“组织学习敏捷度”，同时需在 AI 治理与算法无法替代的人际软技能上加倍投入，推动人类向价值链顶端跃升。 ## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=4s) 权力铁三角到“组织菱形”：人机协同重塑劳动力战略 主持人 Jana Brown 博士以轻松幽默的语调开场。她笑称自己很高兴能接在财务议题之后发言。在传统企业管理结构中，核心决策层往往是由首席执行官（CEO）、人力资源（HR）和财务（Finance）构成的“稳定铁三角”。然而，随着机器深度介入工作流程，“人机协同”（Humans and Machines）正在颠覆这一传统范式，将原本的三足鼎立扩展为容纳技术与机器维度的“四角菱形”（Diamond）。企业当前面临的核心挑战，正是如何让人与机器有机结合，以及这种协同将对未来的整体劳动力战略产生怎样深远的影响。 Jana Brown 博士不仅是一位现任企业首席人力资源官，同时兼具学者与商业高管的双重身份，长期专注于数字化转型背景下的组织变革与学习机制研究。她指出，本次圆桌论坛邀请到的嘉宾阵容完整覆盖了员工生命周期（Employee Life Cycle）的各个切面，旨在从不同视角切入，共同探讨在当今剧烈震化的劳动力市场中，组织究竟该如何重新定义和审视人才战略。 ```mermaid flowchart TD accTitle: 组织治理结构演进：从铁三角到组织菱形 subgraph OldModel["传统管理铁三角"] CEO1["CEO / 最高决策层"] --- HR1["人力资源 (HR)"] HR1 --- FIN1["财务 (Finance)"] FIN1 --- CEO1 end subgraph NewModel["人机协同时代的组织菱形"] CEO2["CEO / 最高决策层"] --- HR2["人力资源 (HR)"] CEO2 --- TECH["机器与 AI 能力"] FIN2["财务 (Finance)"] --- HR2 FIN2 --- TECH HR2 --- TECH end ``` --- ### [01:31](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=91s) 嘉宾阵容亮相：在人机时代重新思考劳动力战略 圆桌嘉宾依次介绍了各自的背景，并分享了当下他们对于劳动力战略最具颠覆性的思考。来自马来西亚最大银行 Maybank（马来亚银行）的前高管 Shameem Farouk 首先发言。他此前长期负责统领该行在东南亚地区覆盖 40,000 名员工的数字化转型历程，历时八至十年，近期刚刚转入学术界。在离开银行业之际，他最为关注的命题是“如何为组织重新找回意义感”。由于 Maybank 的核心使命是“让金融服务更具人性温度”（Humanize financial services），在技术不断自动化的当下，他们激进反思的核心在于如何将使命、内在动机、归属感重新注入到员工向客户提供的核心价值——即深度咨询与高同理心的人际连接之中。 随后发言的是来自培生集团（Pearson）负责组织发展的副总裁 Esin Cooper。她致力于让培生适应 AI 驱动的新工作范式，重塑技能结构与员工的学习、成长及绩效评估机制。Esin 指出，劳动力战略的定义已经发生了根本性转移：在几年前，劳动力战略的黄金法则是“在正确的时间将合适的人放在正确的岗位上”；而今天，战略的本质已演变为“围绕工作本身，将人类能力、AI 能力与学习能力进行最优组合与协同设计，以此创造最大价值”。 紧接着，AI 智能体创业公司 Humaa 的联合创始人兼首席执行官 Giovanni Luperti 分享了自己的见解。Humaa 专门为企业 HR 与 People 团队打造贯穿入职（Onboarding）到离职（Offboarding）全生命周期的 AI Agent。Giovanni 观察到，旧有的人力资源运作逻辑正在迅速失效——管理层级该如何构建、人员编制结构（Headcount Composition）应如何分配，乃至员工体验该如何维护，过去行之有效的假设在当下都必须全盘重新考量。最后登台的 Rosie Buller（来自摩根大通 JPMorgan Chase，担任人工智能与机器学习团队的招聘副总裁）开玩笑说自己坐了一把单独的椅子，“显得格外特别或有些多余”，引得全场会心一笑。她表示非常期待探讨招聘前线中人与 AI 的真实互动与融合。 --- ### [04:50](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=290s) 高管期望转变：劳动力战略即业务战略 Jana Brown 将讨论引向业务领导者（Business Leaders）对 HR 部门不断升级的期望。随着企业由外向内承压，业务线高管已不再满足于传统的人事支持，而是要求 HR 拿出截然不同的战略作为。在银行业深耕数字化多年的 Shameem Farouk 对此感受尤为深刻。 Shameem 直言，现在 CEO 和高管团队（C-suite）对 HR 的要求，是必须彻底重构劳动力战略——将原本基于“照常营运”（BAU, Business as Usual）的日常事务性模式，升级为真正能够兑现 AI 所承诺的生产力跃升的战略引擎。然而，许多管理者习惯将人事与业务割裂，Shameem 则反复向 CEO 和高管层灌输一个根本理念：“劳动力战略就是业务战略，二者绝不能割裂”。 以金融机构的核心价值链为例，越是位于金字塔顶端的高价值客户，其需求往往越集中于高度个性化、复杂的专业咨询服务。这些高价值环节的核心壁垒，完全依赖于资深员工沉淀多年的专业知识与“隐性知识”（Tacit Knowledge）。Shameem 强调，目前没有任何一款 AI 能够真正挖掘并复刻人类这种高度非结构化的隐性经验与专业判断。因此，HR 手中掌握的人才洞察，恰恰是企业参与商业竞争的底牌与核心胜任力。只有确立了“以人才优势赋能高价值业务”的商业顶层设计，招聘等传统功能性事务才会顺理成章地发挥效力。 - **业务战略与人才战略的不可分割性**：企业高价值链条（如高端金融咨询）依赖顶尖人才的不可替代性； - **AI 难以企及的人类壁垒**：员工长期积累的行业洞察、高维同理心与隐性知识（Tacit Knowledge）； - **HR 角色的跃迁**：从被动的日常事务执行者，转变为利用人才画像定义商业竞争优势的战略伙伴。 --- ### [08:11](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=491s) 组织发展新范式：从“适应系统”到“重构人机系统” 针对培生集团（Pearson）组织发展副总裁 Esin Cooper 的工作，Jana 评价其代表了 HR 领域中“最具体系化（Systemic）的硬核工作”。Esin 深入剖析了业务部门对 OD（组织发展）诉求的前后剧烈反差。 过去，业务主管们往往在项目已经推进到一半时才敲开 HR 的门，直接下达指令，要求 HR 配合完成既定方案。而今天，高管们带来的问题变得极其具有战略前瞻性：“我们到底应该如何重新组织工作？”、“哪些工作必须由人类主导？哪些环节可以通过 AI 进行增强（Augmented）？又有哪些必须实现全自动化？”以及“在这样一个全新时代，我们的领导者究竟需要具备什么样的能力来带领团队？” 面对这些提问，Esin 表示培生目前正以 AI 为核心透镜，与业务领导者展开紧密协作，重塑整体“职业架构”（Career Architecture），并对所有岗位与工作流进行逐一拆解与推敲。整个组织发展的重心发生了一场标志性转向：过去是“帮助员工去适应既定系统”（Help people adapt to the system），而现在则是“帮助企业彻底重构系统，让员工能与 AI 并肩协作并取得成功”。这一转变的核心在于精准界定人机分工——在需要人类高阶判断力（Human Judgment）的关键链条中坚守“人类在回路中”（Human in the loop），而在重复与辅助环节最大化释放机器生产力。 --- ### [09:55](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=595s) 招聘与人资演进：AI 适应力与“HR 的终结” 话题随后切入人才选拔的最前线。摩根大通的 AI/ML 招聘副总裁 Rosie Buller 从用人主管与求职候选人两个视角剖析了招聘逻辑的蜕变。在用人端，面试官不再仅仅关注候选人会操作哪些具体的 AI 工具，而是深入考察其对 AI 的“适应力”（Adaptability to AI）——即候选人能否利用技术真正撬动业务影响力。在求职端，传统的两页静态简历已经无法展现一个人的真实水平。候选人必须阐明自己是如何将 AI 工具与人类专业能力进行深度结合的，最关键的是证明自己“不会盲信或完全依赖工具输出的结果”，而是能够在结果落地时注入人类独有的判断力。Rosie 坚信，无论技术如何迭代，人与人交流中流淌的直觉与价值裁决，是 AI 永远无法削弱的核心桥梁。 ```mermaid flowchart LR accTitle: 人才招聘评估维度的范式转变 subgraph Traditional["传统招聘考核维度"] T1["静态两页简历"] --> T2["特定工具/技能掌握度"] T2 --> T3["单兵独立产出能力"] end subgraph Modern["AI 时代招聘考核维度"] M1["AI 适应力与批判思维"] --> M2["人机整合与业务落地能力"] M2 --> M3["不可替代的人性判断与直觉"] end ``` 谈及行业大势，Jana 引出了 Humaa 公司极具冲击力的品牌口号——“HR 的终结”（The end of HR）。Humaa CEO Giovanni Luperti 对此做出了解释：这并非意味着人力资源职能的消亡，而是“我们所熟知的传统 HR 模式的终结”（The end of HR as we know it）。 Giovanni 指出，在企业实际落地中，HR 团队的技术化倾向日益显著。一方面，首席人力资源官（CHRO/CPO）正前所未有地与首席技术官（CTO）和首席信息官（CIO）深度捆绑，共同搭建贯穿员工全生命周期的底层 AI 基础设施；另一方面，工程研发人员、数据分析师等硬核技术人才开始大规模直接入编 HR 团队。这种结构性变革直接撼动了传统的组织假设：企业是否还需要像过去那样配备庞大基数的“HR 业务合作伙伴”（HRBP）？抑或是通过技术赋能与 AI Agent 架构，以更轻量、敏捷、高速的全新形态驱动组织扩张？传统 HR 的游戏规则正在被全面重写。 ### [13:33](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=813s) 招聘中的“人机协同”：为何简历筛选仍需要 Human-in-the-Loop 讨论焦点转向了员工生命周期（Employee Lifecycle）。主持人和嘉宾们一致认为，员工体验（EX）越好，客户体验（CX）就越好。当探讨如何将 AI 融入招聘流程并保持人本核心时，负责 AI/ML 人才招聘的嘉宾 Rosie 指出，尽管现代 ATS（申请人跟踪系统）深度集成了各类 AI 自动化功能，但在审阅候选人简历时，依然必须坚持 "Human-in-the-loop"（人在回路中）。 Rosie 举了一个生动的行业痛点案例：当技术主管需要寻找具备某种特定 AI 细分领域专长的人才时，候选人很难把所有细节压缩在短短两页纸的简历中。然而，经验丰富的人工招聘团队会敏锐地注意到简历上的学术出版物或专利线索，主动前往 Google Scholar（谷歌学术）等学术平台深挖，从而证实候选人曾在博士阶段或学术论文中深入研发过该技术。如果完全依赖 AI 关键词自动筛选，这类顶级顶尖人才就会被直接漏掉。“我们不可能让候选人写六页纸的简历，但如果只看两页纸且完全交由 AI 筛选，我们会错失大量优秀人才。” ```mermaid flowchart LR accTitle: 人机协同简历筛选流程 subgraph S1["传统 AI 筛选局限"] A["2页纸标准简历"] --> B["ATS / AI 自动关键词过滤"] B --> C["遗漏深层专利与学术成果"] end subgraph S2["Human-in-the-Loop 增效"] A --> D["人工招聘专员深度复核"] D --> E["拓展查询 Google Scholar / 专利库"] E --> F["成功识别稀缺 AI/ML 顶级专家"] end ``` --- ### [15:48](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=948s) 从断点式培训到持续能力建设：培生（Pearson）的“核心动力技能” 在谈及组织发展（OD）的演进时，培生（Pearson）组织设计与人才供应链副总裁 Esin Cooper 表示，无论是培生内部还是其企业级客户，人才培养模式正在发生根本性变革：从过去的“断点式/偶发式培训”（episodic development）转向“持续性能力建设”（continuous capability building）。传统的入职培训与技能晋升项目依然有其价值，但当下最核心的命题是：如何帮助员工以“跟得上工作变革的速度”去加速学习。 为了应对这一挑战，培生去年推出了全新的“培生动力技能”（Pearson Power Skills），重点聚焦于三大支柱：**学会如何学习（Learning to learn）**、**适应力（Adaptability）** 以及 **AI 熟练应用度（AI Proficiency）**。Esin Cooper 将这些技能定义为推动业务发展的“力量倍增器”（Force Multiplier），旨在帮助员工在日常工作中快速构建新能力、从容应对剧烈变革，并将 AI 以最相关、最高效的方式深度融入日常业务。 - **模式转变**：从阶段性入职培训转向伴随业务变化的实时、持续性学习。 - **培生三大动力技能**： - **学会如何学习**：培养员工自主敏捷获取新知识的能力。 - **动态适应力**：提升应对不确定性与业务转型的抗压与调整能力。 - **AI 实际应用度**：将生成式与自动化 AI 工具转化为日常工作的高效杠杆。 --- ### [16:59](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1019s) 敏捷与使命驱动：马来亚银行三周上线电商平台的启示 人力资源如何避免与一线业务脱节？马来亚银行（Maybank）集团科技、数字银行与数据人力资本总监 Shameem Farouk 指出，HR 当前最艰巨也最核心的任务在于**组织设计与文化塑造**。他以马来西亚最大银行在疫情期间的真实案例说明了使命感与打破教条的巨大威力。疫情初起时，大量中小企业（SME）因无法正常开门营业而面临倒闭危机。 面对国家经济与社会的严峻危机，马来亚银行一支跨部门团队在强烈的社会责任感与使命感驱使下迅速集结，在短短 **3 周** 内就自研并上线了一套全新的中小企业电商平台。对于一家拥有 60 余年历史的传统大型金融机构而言，3 周内完成产品研发并推向市场是“闻所未闻的奇迹”。这一创举之所以能够实现，关键在于团队在非常时期彻底打破了僵化的“职位描述”（Job Description）： > “没有死板的职位描述，因为在那种特殊时期没人能预料未来。真正驱动员工的是使命感、奋斗目标以及对客户的服务意识。” Shameem 强调，面对新一轮 AI 技术变革浪潮，HR 必须推动管理层和招聘团队转变思维：**不再按静态任务（Task）与固定 JD 招人，而是基于价值观、持久性底层技能（Enduring Skills）、批判性思维、复杂问题解决能力以及跨学科综合视角来进行人才选拔与组织设计。** --- ### [19:57](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1197s) 4万亿美元AI投资浪潮下的 ROI 衡量与 Agent 规模化扩展 面对全球预计高达 4 万亿美元的先锋技术与 AI 投资预算，企业高管究竟该如何衡量技术在全企业推广落地中的投资回报率（ROI）？Humaa 首席执行官 Giovanni Luperti 分析指出，领军企业目前主要通过两大核心维度来衡量 ROI： 1. **运营效率（Operational Efficiency）**：企业期望在不按比例大幅扩张员工编制（Headcount）的前提下实现业务规模化扩张。AI 在此充当了个人产能的“杠杆乘数”（Multiplier），企业需要精准识别出哪些业务节点可以通过该乘数实现倍增，并对这些领域进行激进投资。 2. **卓越绩效表现（Performance Impact）**：通过剥离高绩效核心人才日常承担的低价值杂务与冗余交互，使他们能将全部精力集中在驱动组织核心业务价值上。 Giovanni 还提出了一个关于 **AI Agent（智能体）** 的重要视角：从扩展工作规模的角度来看，AI Agent 与人类员工非常相似，但其在特定垂直任务上的**扩展速度（Velocity）**要快得多。企业可以像规划人力编制一样去规划 Agent 的扩容，但在执行特定任务时，Agent 的边际成本仅为人力的一小部分。 --- ### [21:53](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1313s) 顶尖企业的 AI 落地秘诀与常见陷阱：告别“打卡式试点”与重视变革管理 谈及头部企业的“成功秘诀”，Giovanni Luperti 强调，顶尖组织在部署 AI 能力时具有极强的**明确目标性（Incredibly Intentional）**。与之相对，许多挣扎中的企业仅仅把 AI 当作一种“打卡清单式练习”（Checkbox Exercise）——为了对外宣称“我们在投资 AI、我们在思考 AI”而零散地做几个浅层试点，最终收效甚微。 优秀的实践范式是选择一个极其具体的高价值场景进行深度攻坚。例如，彻底重构并自动化整个“员工入职（Onboarding）流程”，在用户体验、流转速度与自动化程度上做到极致，并在团队重构、技术适配与**变革管理（Change Management）**上投入充足资源。 | 评估维度 | 成功企业的做法（激进且深度） | 挣扎企业的通病（浅层且盲目） | |:--- |:--- |:--- | | **战略意图** | 目标明确，直击具体业务痛点（如全流程入职自动化） | 为了 AI 而做 AI，仅当成“打卡式”试点项目 | | **技术视角** | 业务导向：清晰定义 AI 部署的终极业务目标 | 唯技术论：过度沉迷于技术本身，脱离实际场景 | | **组织与变革** | 深度重组团队、明确治理架构与权责下放机制 | 严重低估变革管理难度，维持旧有组织架构不愿改变 | Giovanni 直言，组织受阻的最大根源往往在于**“误以为自己想要 AI，但实际上根本没做好迎接组织变革的准备”**。 --- ### [23:56](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1436s) 破解组织内部阻力：跨职能特别行动组（Task Force）与敏捷重塑 针对 Giovanni 提出的变革管理挑战，Shameem Farouk 深表赞同并分享了数字化落地中的内部博弈。他透露，在推进组织数字化与智能化转型时，最出人意料的阻力往往来自**技术部门自身**以及**业务一线的惯性思维**： - **技术部门的顾虑**：IT 运维团队背负着系统 100% 正常运行时间（Uptime）与系统稳定交付的严苛 KPI，任何可能导致故障的试验和失败都是他们极力排斥的。 - **业务团队的惯性**：全员都在忙于维持现有业务的运转，在人人超负荷的情况下，“谁有空来创造变革？” ```mermaid flowchart TD accTitle: 破解组织阻力的跨职能特别行动组模式 subgraph InternalFriction["双重内部阻力"] T["技术部门<br/>(严控故障率/抗拒试错)"] B["业务一线<br/>(忙于日常运转/习惯路径)"] end InternalFriction --> P["设立跨职能敏捷特别行动组<br/>(Cross-functional Task Force)"] subgraph Solution["破局机制"] P --> D1["从各业务线抽调核心人才"] P --> D2["作为独立专项项目运作"] P --> D3["重塑组织设计并实施敏捷办公"] end ``` 为了打破这一僵局，马来亚银行不再依赖自上而下的单一技术指令，而是成立了由各业务线骨干组成的**跨职能特别行动组（Cross-functional Task Force）**。该行动组作为独立专项项目运作，重新审视整体组织设计，构建敏捷工作环境。Shameem 总结道，自动化与 AI 落地绝不能仅仅从纯技术视角切入，必须从业务、技术与人性的多维视角协同推进，才能彻底激活变革。 ### [25:56](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1556s) 组织变革与重塑：激励机制、跨职能协作与数据技术底座的民主化 在推动组织深层变革时，跨职能团队的领导架构与参与机制至关重要。Shameem Farouk 结合马来亚银行（Maybank）的实际经验指出，驱动员工行为转变的核心杠杆在于重构“工作激励”。这里的激励并非单纯指绑定数字化指标的奖金，而是非物质性的认可与提拔——重点晋升那些积极拥抱数字化变革、主动发起改变的“先锋领袖”。变革工作流的领导权并不拘泥于技术部门、业务部门或人力资源部，而是由体现组织核心领导力价值观的人选来牵头。由于入选该专责小组被视为极高的荣誉，员工们形成了“我必须参与其中”的积极氛围；选拔领导者的标准也不再局限于技术专长，更看重其跨职能沟通、协作以及优秀的团队管理能力。 与此同时，作为一家拥有 60 多年历史的老牌银行，组织内部不可避免地存在严重的部门壁垒。Shameem 强调，要运行真正高效的机器学习算法，绝不能仅局限于单一职能视角，算法必须跨越组织边界、调用来自不同业务板块的多维变量。因此，团队经历了反复沟通与拉锯，推动技术部门实现技术栈与底层数据的“民主化”（开放与共享）。在强监管和严格数据隐私保护要求的金融背景下，打破数据壁垒、建立开放的基础设施是一项极其艰巨的挑战。Shameem 总结道，重塑激励机制、技术与数据民主化、组织架构设计是不可或缺的三大基石；如果这三者未能就位，单靠再培训（Reskilling）根本无法奏效，因为员工即便掌握了新技能，也缺乏在实际工作中应用的动力与环境。 ```mermaid flowchart TD accTitle: 组织再培训与数字化转型的三大核心支撑 subgraph Pillars ["转型三大核心支柱"] A["激励机制重构<br/>（非物质认可、提拔变革先锋）"] B["技术与数据民主化<br/>（打破部门孤岛、跨业务数据共享）"] C["组织架构再设计<br/>（跨职能协作与流程重构）"] end Pillars --> D["有效的员工技能再培训<br/>（Reskilling）"] D --> E["业务落地与行为持续转变"] ``` - **非物质激励导向**：通过提拔跨职能变革拥护者树立榜样，激发全员参与专项任务小组的内驱力。 - **跨部门算法需求**：机器学习模型的效能依赖于全局多维数据，倒逼传统金融机构开放底层基础设施。 - **技能重塑前提**：激励机制、数据底座、组织设计三者联动，是确保员工技能培训转化为实际生产力的必要条件。 --- ### [30:08](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1808s) 克服适应性障碍：将 AI 打造为解决业务痛点的“止痛药” 业界数据显示约有 70% 的 AI 项目未能达到预期目标。针对组织适应性障碍，Esin Cooper 指出，最普遍的误区是将 AI 单纯视为“技术部署”。技术本身的落地相对容易，但要改变员工的工作习惯、重塑决策机制以及调整岗位价值定义却极其困难。另一个常见痛点是人力资源、技术与业务部门步调不一致——各方都对 AI 充满热情，却各自推进独立的变革日程、时间表和交付承诺。AI 转型本质上是全员协同的“劳动力转型”，而非孤立的技术战役。 此外，领导层若无法做到“以身作则”（Role Model），整天空谈 AI 却不主动学习、实验或将其融入日常，团队便很难真正跟进。Esin 强调，要推动全员采纳，必须将 AI 深度嵌入日常业务节奏中，让它成为解决繁琐流程的“止痛药”（Painkiller）。以培生集团（Pearson）的绩效管理为例，员工往往忙于业务交付与销售业绩，视绩效考核为行政负担。为此，团队开发了一款名为“Bar Up”的 AI Agent，协助员工和管理者制定清晰智能的 OKR、进行全年绩效回顾并撰写评估初稿，从而将管理层从繁琐的案头工作中解放出来，专注于高价值、有深度的面对面沟通。 - **70% 失败率根因**：孤立看待技术部署，忽视了工作方式与决策流程的人性化重塑。 - **三方协同脱节**：HR、IT 与业务部门各行其是，缺乏统一的劳动力转型战略路线图。 - **“止痛药”定位**：培生推出“Bar Up”AI Agent，自动化处理绩效管理的行政事务，赋能管理者聚焦有意义的对话。 --- ### [33:29](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2009s) 学习先行与自下而上的赋能：以招聘场景为例的技术降维理解 在探讨 AI 落地路径时，Rosie 提出组织应确立“学习优先”的原则。无论系统集成的能力有多强大，如果员工无法理解和使用，工具的价值就会归零。因此，企业需要开展密集的学习培训课程，从基层到管理层全面培育 AI 应用能力，并在员工职业生涯的各个阶段持续评估使用情况并提供学习支持。 Rosie 以招聘团队为例展示了 AI 的降维赋能效果：以往招聘人员（Sourcers/Recruiters）在面对复杂的工程技术术语时往往难以深入理解，而现在借助生成式 AI，他们可以让系统用“面对幼儿园小朋友般的通俗语言”将高深的技术栈层层拆解。这种微小而强大的应用不仅消除了业务沟通壁垒，让招聘人员能够精准把握用人经理的技术需求，也生动证明了自下而上推广 AI 学习与工具使用的巨大潜力。 - **学习先于集成**：AI 工具的实际利用率取决于全员的学习适应能力，必须建立贯穿职业周期的培训支持。 - **招聘场景技术降维**：利用 AI 将晦涩的工程概念翻译为通俗语言，大幅降低非技术人员理解前沿技术的门槛。 --- ### [35:25](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2125s) AI 时代的领导力重塑：从“无所不知”到“终身学习”的榜样力量 谈及 AI 时代领导者所需的关键转变，Shameem Farouk 强调了“展示学习能力与展现脆弱（Vulnerability）”的重要性。回溯到 2017 至 2018 年机器学习刚兴起时，马来亚银行的首席执行官、董事会及全体高管团队特意抽出两天半的时间参加闭门学习研讨会。最高管理层从零开始探索算法原理，甚至亲自上手学习编写 Python 代码。人力资源团队将高管们敲代码的照片在全行公开分享，向全员传递出明确信号：最高层愿意承认“我不知道”，并主动将心态从“无所不知”（Know-it-all）转变为“终身学习”（Learn-it-all）。这种打破完美假象的举动极大地消除了普通员工面对新技术时的恐惧感与“冒名顶替综合征”（Impostor Syndrome），在全行确立了允许试错、开放包容的创新文化。Shameem 补充道，在技术普惠的今天，最顶尖的解决方案完全可能来自一名实习生，管理者必须保持谦逊与开放。 ```mermaid flowchart LR accTitle: 领导力心态转变对组织文化的影响 subgraph MindsetShift ["领导者心态转变"] A["'无所不知'<br/>（Know-it-all）"] -->|展现脆弱与学习榜样| B["'终身学习'<br/>（Learn-it-all）"] end subgraph CulturalImpact ["组织文化效应"] B --> C["消除冒名顶替综合征"] B --> D["包容试错与心理安全感"] B --> E["打破层级倾听一线创新<br/>（如采纳实习生建议）"] end ``` Esin Cooper 对此高度赞同，并指出面对 AI 带来的颠覆性变革，领导者需要具备更强的适应力、韧性与敏捷行动力。在培生集团，除了工程研发等纯技术岗位外，高级管理者被明确要求在 AI 相关的核心能力（Power Skills）上达到最高熟练度等级。领导者必须深刻理解技术如何重塑自身业务领域，以无畏的姿态走到变革前沿，成为拥抱学习文化的先锋表率。 - **高管亲自编程**：马来亚银行 C-suite 展开 2.5 天 Python 研讨会，以“展示脆弱”破除全员技术畏难情绪。 - **心态模式更迭**：领导力核心从展现权威转变为展现“求知与包容失败”，打破层级吸纳全员智慧。 - **高管能力硬性要求**：培生将最高等级的 AI 熟练度要求直接绑定至资深管理层，倒逼领导层保持前瞻与敏捷。 ### [38:38](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2318s) 领导者在未知中的导航与跨代协同 在深入探讨组织变革与持续学习之后，嘉宾们聚焦于当下管理层所面临的核心挑战：应对极端的不确定性。Giovanni Luperti 强调，今天的管理者必须在心理上对“模糊性”（ambiguity）感到极度适应，并学会如何在未知中导航。过去，领导者习惯于依赖同行验证与大量的社会认同（social proof）来做出决策，但生成式 AI 技术的发展日新月异——仅仅在两年前，它甚至还没有达到如今这样的普及与影响程度。目前业内缺乏现成的成熟经验与标准模板，没有人能确切地告知每一步行动的具体成效。因此，领导者必须学会在某种程度的“不适感”中前行，秉持“快速行动并勇于打破常规”（move fast and break a few things）的精神，去探索未知的机遇。 与会嘉宾随后从团队协作的角度补充了对领导力的要求。领导者不能继续留在高高在上的孤岛中，而应当深入基层一线，花费更多时间与年轻工程师、实习生等团队成员交流。年轻一代往往能更快掌握最新的 AI 工具与应用方式，管理层可以从下属身上学到很多东西。AI 时代的到来需要跨代际的深度协同，既要发挥 AI 的效率，也要保留人际互动的温情，将两者的优势结合起来。 ```mermaid flowchart TD accTitle: AI时代领导力转型模型 accDescription: 展示领导者如何从传统的孤岛式管理转向拥抱模糊性与一线协同。 subgraph Mindset["思维模式重塑"] A["拥抱模糊性与不适感"] --> B["快速行动与迭代试验"] end subgraph Collaboration["跨代际协同"] C["打破管理孤岛"] --> D["深入一线向年轻员工/实习生学习"] D --> E["融合技术效能与人类温情 (Human Touch)"] end B --> Collaboration ``` - **缺乏先验经验**：生成式 AI 发展极快，企业无法等待市场上出现完全成熟的“社会认同”后再行动，所有组织都处于同步试错与摸索阶段。 - **向下学习（Reverse Mentoring）**：新生代员工与初级工程师对前沿工具更具敏感度，管理层需要打破层级壁垒，促进双向知识流动。 --- ### [40:14](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2414s) AI 时代繁荣企业的核心特质：启蒙智能与人本价值 主持人向全场嘉宾抛出一个句子填空题：“在 AI 时代能够蓬勃发展的组织，将会是那些能够______的组织。” 各位嘉宾从不同维度给出了精辟的见解： - **坚守人类温度与互助学习**：企业是由人构成的，员工不是冰冷的机器人，大家以不同的节奏成长，唯有保持“人性的温度”（human touch）并彼此赋能，组织才能长盛不衰。 - **敏捷应变与快速推进**：Giovanni Luperti 简明扼要地总结为“快速行动并拥抱变化”。 - **同等重视技术与人力资产**：Shameem Farouk 提出，企业在部署 AI 技术时投入了多少心力与资源，就应当以同等强度的意图（intentionality）去构建人类的能力。 - **将人工智能转化为“启蒙智能”（Enlightened Intelligence）**：这一理念借鉴自印第安纳大学研究学者的观点，强调企业不能仅仅停留在纯技术的 AI 应用上，更要引导员工持续培育意义感、使命感与价值观，以负责任的方式推动人类与科技共同繁荣。 --- ### [41:40](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2500s) 决胜未来：企业当下必须构建的关键能力 针对“企业现在最需要立即着手构建的一项能力是什么”这一问题，嘉宾们从价值链升级、组织敏捷度、人际技能以及治理等多个维度展开了探讨： ```mermaid flowchart LR accTitle: AI时代核心能力构建框架 accDescription: 总结嘉宾提出的四项关键组织能力。 A["AI时代核心能力"] --> B["推动人类向价值链顶端跃升<br/>(提升能力/做大蛋糕)"] A --> C["组织学习敏捷度<br/>(洞察趋势/快速转化业务成果)"] A --> D["强化不可替代的软技能<br/>(聚焦人际技能与AI高价值落地)"] A --> E["建立全方位AI治理机制<br/>(保障合规/伦理与安全)"] ``` 嘉宾们认为，首要能力在于探索如何真正“将人类推向价值链的高端”（move humans up the value chain），利用技术增强人类能力，创造共享繁荣，把业务与价值的“蛋糕做大”。其次，核心竞争优势将直接源自**组织的学习敏捷度（organizational learning agility）**——能够洞察未来趋势、快速识别所需的新型人才能力，并以最快速度完成能力构建并应用于实际业务，从而驱动切实的商业成果与影响。与此同时，企业既要找准 AI 能产生最大价值的业务切入点，更要在不可被算法轻易替代的**人际技能（people skills）**上加倍投入。最后，严密健全的**治理（governance）**体系是不可或缺的底线保障，企业绝不能忽视 AI 的合规与伦理机制。 - **做大价值蛋糕**：技术应用的核心目标不是压缩人力成本，而是通过赋能人类创造更大的整体商业与社会价值。 - **学习敏捷度决定胜负**：敏锐洞察技能缺口并将其迅速转化为业务落地的组织，将建立长期的竞争壁垒。 - **软实力投资**：技术越普及，组织越需要加倍投资于那些无法被机器替代的情感、沟通与人际协作能力。 --- ### [43:36](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2616s) 总结与跨部门联盟建设 在讨论的收尾阶段，主持人总结了整场论坛的核心主旨：HR 绝不能在真空中推进组织变革。从企业文化、人才选拔、领导力发展到整体组织架构，人力资源部门必须与 CEO、首席财务官（CFO）、技术领袖以及各业务板块负责人紧密绑定、结为同盟。唯有建立起跨职能的强大联盟，组织才能在 AI 浪潮中保持敏捷、高效推进全面变革，并在未来的智能时代中立于不败之地。 ## Notable quotes > [01:31](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=91s) 围绕工作本身，将人类能力、AI 能力与学习能力进行最优组合与协同设计，以此创造最大价值 — Esin Cooper > [04:50](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=290s) 劳动力战略就是业务战略，二者绝不能割裂 — Shameem Farouk > [08:11](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=491s) 帮助企业彻底重构系统，让员工能与 AI 并肩协作并取得成功 — Esin Cooper > [13:33](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=813s) 我们不可能让候选人写六页纸的简历，但如果只看两页纸且完全交由 AI 筛选，我们会错失大量优秀人才。 — Rosie Buller > [16:59](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1019s) 没有死板的职位描述，因为在那种特殊时期没人能预料未来。真正驱动员工的是使命感、奋斗目标以及对客户的服务意识。 — Shameem Farouk > [21:53](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=1313s) 误以为自己想要 AI，但实际上根本没做好迎接组织变革的准备 — Giovanni Luperti > [38:38](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2318s) 快速行动并勇于打破常规 — Giovanni Luperti > [40:14](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2414s) 企业在部署 AI 技术时投入了多少心力与资源，就应当以同等强度的意图（intentionality）去构建人类的能力。 — Shameem Farouk > [41:40](https://www.youtube.com/watch?v=TbaTE7oV7eI&t=2500s) 将人类推向价值链的高端 — Esin Cooper ## Mentioned Jana Brown, Esin Cooper, Shameem Farouk, Giovanni Luperti, Rosie Buller, Pearson (培生集团), Maybank (马来亚银行), eHealth, Humaa, JPMorgan Chase (摩根大通), Google Scholar, Python, Bar Up, Indiana University (印第安纳大学) ## Video description Esin Cooper - VP, Organization Design & Talent Supply Chain, Pearson Shameem Farouk - Human Capital Director, Group Technology, Digital Banking, and Data, Maybank Jana Brown - Chief Human Resources Officer, eHealth Giovanni Luperti - CEO, Humaa As AI automates tasks and augments decision-making, HR leaders must rethink everything from hiring to career development. This panel will explore how organizations are using AI to identify skills gaps, personalize learning paths, and enhance employee experiences. We’ll also tackle hard questions about surveillance, bias, and fairness in people analytics. Finally, we will address the HR function and role of human workers as AI agents become commonplace and carry out more work. Learn how to balance innovation with empathy—and build a workforce that thrives in an AI-augmented future [Source](https://www.youtube.com/watch?v=TbaTE7oV7eI)

[1:31] Esin Cooper: 围绕工作本身，将人类能力、AI 能力与学习能力进行最优组合与协同设计，以此创造最大价值 [4:50] Shameem Farouk: 劳动力战略就是业务战略，二者绝不能割裂 [8:11] Esin Cooper: 帮助企业彻底重构系统，让员工能与 AI 并肩协作并取得成功 [13:33] Rosie Buller: 我们不可能让候选人写六页纸的简历，但如果只看两页纸且完全交由 AI 筛选，我们会错失大量优秀人才。 [16:59] Shameem Farouk: 没有死板的职位描述，因为在那种特殊时期没人能预料未来。真正驱动员工的是使命感、奋斗目标以及对客户的服务意识。 [21:53] Giovanni Luperti: 误以为自己想要 AI，但实际上根本没做好迎接组织变革的准备 [38:38] Giovanni Luperti: 快速行动并勇于打破常规 [40:14] Shameem Farouk: 企业在部署 AI 技术时投入了多少心力与资源，就应当以同等强度的意图（intentionality）去构建人类的能力。 [41:40] Esin Cooper: 将人类推向价值链的高端

[0:04] Shameem Farouk: restoring meaning, purpose, and motivation to core advisory services to humanize financial services [0:04] Esin Cooper: simply placing the right people in the right roles at the right time to designing an optimal combination of human, AI, and learning capabilities around work to drive value [0:04] Giovanni Luperti: traditional assumptions regarding management layers, headcount composition, and employee experience are rapidly becoming obsolete [4:50] Shameem Farouk: workforce strategy cannot be divorced from business strategy [13:54] Esin Cooper: Pearson Power Skills [30:31] Esin Cooper: Bar Up [33:54] Shameem Farouk: know-it-all [33:54] Shameem Farouk: learn-it-all [37:35] Shameem Farouk: enlightened intelligence

[0:04] Hello everyone. So glad to see you all here. Uh  we have a very important topic and I uh can just [0:12] say that I'm glad to be following finance. As I  think about the triangles in organizations, it's [0:20] typically the CEO, HR, and finance. Now that's  changing and a lot of it's changing because of [0:29] the topic we're talking about today is becoming a  diamond particularly related to having humans and [0:38] machines coming together and how do we make that  work and what are the implications to workforce [0:45] strategy. That's what we're talking about today  and boy do we have an incredible panel. I'm Dr. [0:51] Janna Brown. In my day job, I'm a chief human  resources officer. I consider myself a business [0:57] executive and an academic. I have studied uh  learning and organizational change specifically [1:06] in digital transformation. We have the entirety  of the employee life cycle represented here. Uh [1:14] so what we're going to start off doing is having  some introductions. So let's jump right into it. [1:21] Uh introduce yourselves and then also share what  you're rethinking today as it relates to workforce [1:31] strategy. Hi uh good afternoon everyone. So my  name is Shamim. Uh so I actually was previously in [1:41] a financial institution called May Bank uh a bank  in Malaysia. I just left the bank actually. So I'm [1:48] uh gone into academia now but I was responsible  for digitalizing the workforce of 40,000 employees [1:56] across Southeast Asia and we have been going  through this journey for the past 8 10 years. [2:03] So I would be happy to share their experiences  and what I think before I left what we were [2:10] radically thinking is uh the whole area on  how do we preserve meaning back uh to the [2:17] organization because our mission is to humanize  financial services and uh we believe that at the [2:26] crux of it is bringing back meaning, purpose uh  motivation, engagement ment to the core service [2:35] that we provide to employees which is advisory and  a very human uh connection to our customers. So [2:43] that's wonderful rethinking the strategy or what  you're focusing on is bringing meaning back. Got [2:50] it. Love it. Please. Right. Hi everyone. I'm Essen  Cooper and I'm from Pearson. I lead organizational [2:56] development there. Um, and that means that I'm  focused on everything to do with preparing Pearson [3:01] for uh the new world of work as it's transforming  with AI and how skills are reshaping and how our [3:09] people learn, grow, and perform. So, and  that's very much what I'm focused on. Uh, [3:14] a couple of years ago, I would have said workforce  strategy was about getting the right people in the [3:19] right roles at the right time. Um, and now I'd say  it's really about designing the right combination [3:25] of human capability, AI capability, learning  capability around work so that we can create [3:32] value. Love that. Thank you. Thanks very much. Uh,  Joani Luperti, I'm the CEO of humans and we build [3:40] AI agent for HR teams and people teams. We support  companies across the full employee life cycle from [3:47] onboarding to offboarding and anything in between.  and uh we work with enterprise organization to [3:53] kind of you know help them understand how can  rethink a number of processes in the context of [3:59] um of AI. Something that for me personally has  been uh quite fascinating to observe in the [4:03] context of workforce strategy is how everything  is changing incredibly fast and there is a number [4:08] of assumption and things that we observed in  the past that are no longer relevant in today's [4:14] context from like how do you build the management  layer how you think about headc count composition [4:19] and employee experience as well. I'm excited  to talk more about this today. Thank you and [4:24] Rosie. Yeah thank you. I feel like I either I  am special or extra. I got separate chairs. Hi, [4:32] I'm Rosie Buller. I work at JP Morgan Chase. I am  a vice president of AIML uh recruiting team there [4:41] and I'm excited here to talk about human with AI  how it's working. Thank you. So let's focus on [4:50] expectations because as we progress and things  change so much as we think about our clients, [4:58] the business leaders, they're expecting something  more and different from us. So share with us what [5:07] those changes look like. Let's start with you  Shim. You've been engaged in transforming your [5:12] organization for a couple of years now. How are  business leaders expectations of HR changing from [5:19] your perspective? Thank you for that uh question  because I think it's a really important one. [5:25] I think uh my CEO my business leaders right seues  are looking at me uh and the HR function to to [5:35] talk about how do we actually revamp the whole  workforce strategy from one of uh the doing from [5:44] a BAU business as usual perspective to one that  can actually um you know deliver the productivity [5:52] gains that AI is supposed to deliver And so uh the  CEO is asking for a workforce strategy but I think [6:01] uh from my perspective the way we deliver  workforce strategy is to talk about business [6:07] strategy. They are intertwined. So and this is  something which I had to also educate both my HR [6:16] colleagues and my you know functional seuitees  and the CEO that we cannot divorce the two. [6:24] HR workforce strategy is business strategy  and the reason for that is because the [6:30] uh core value chain the value that you provide  for financial institution the higher you go or the [6:38] higher value customers are those customers that  actually require the highest amount of advisory [6:44] services right just as an example where the value  is so from a business competitive perspective [6:52] uh I I think we need to see where exactly  the opportunities are based on the talent [6:58] strengths that you have. the talent provides.  I mean the insights that HR has on the talent [7:04] is the advantages the core competencies that  you can use to deliver the enhanced value back [7:12] because the value that uh our employees have they  have expertise they have tacid knowledge and this [7:18] is what I mean AI cannot easily replicate  I don't know of any AI yet which is able to [7:25] draw out expertise and tacid knowledge right  that can actually provide uh the huge amount [7:32] of value of advisory for example especially at  the higher end to customers and that's where [7:37] the value is for large institutions. So it's it  is about business strategy you know not spoken [7:43] at all about you know hiring or or or all that  the the typical functional stuff it will come [7:48] naturally once you set the strategy right. I love  that. And ensuring that the business leaders are [7:55] always thinking about HR when they're focused on  strategy. Uh because that is going to be the most [8:03] critical way you are able to pull it together not  as an afterthought but together. um you know when [8:11] we look at HR and the services we provide as  in the work that you do is probably the most [8:19] systemic work that we have in HR organizational  development. So what are leaders asking you to do [8:28] differently? Well, previously they come to us when  they were halfway through the process and tell us [8:33] exactly what they needed. Um and now the questions  are far more strategic. So they're coming to [8:38] us and they're saying well how is how should  we organize work and what should be humanled [8:44] versus augmented by AI versus automated and what  capabilities do our leaders need in our workforce [8:51] to lead in this new era. I love those questions.  Um and we are partnering very closely with our [8:57] business leaders now. Together we're really  using the lens of AI to rethink work itself. [9:04] embedding AI in our career architecture and we're  really um examining all of our roles and workflows [9:10] to do exactly that. What needs to remain humanled  with human judgment and then what can we augment [9:16] with AI or automate to create the most value. So  um there's been a real shift in before it was kind [9:24] of help us help people to adapt to the system and  now they're really coming to us and saying help [9:29] us redesign the system so that our people can  work side by side with AI and um be successful. [9:36] That's very encouraging that we are focusing  on how the work is being redesigned because [9:43] it is there that we'll be able to focus on the  human in the loop and then where the machine also [9:50] uh stays with us and as we go we're going to go to  the end there Rosie in the special chair because [9:55] of of me over here. Uh but Rosie, as you who's in  recruiting, uh from a hiring leaders perspective, [10:06] you're working very closely with hiring leaders.  How is AI changing the expectations of what great [10:13] talent looks like? Yeah, that's a great question.  I always see when it's come hiring, there's two [10:22] parties. One is hiring team and is candidate.  So now how with AI with human it's evolving in [10:31] recruiting I see three things in hiring leadership  where they are evaluating more these candidates [10:40] their adaptability to AI I'm not talking about  tools here how they are looking at to use to make [10:49] the workforce impactful so that part is becoming  very critical when leaders are interviewing [10:57] interviewers are happening that part the other  side candidate when they are interviewing I would [11:05] say it's more than just two page of résumés when  we are talking you have to talk not just about [11:14] the tools of AI somebody is using but how it's  making an impact you are not just depending your [11:23] decisions what it's coming out of those tools how  AI and human get integrated right there at the end [11:31] of the uh results. So that part is very critical  when somebody is talking with the hiring leaders [11:38] and there's a beautiful bridge between these two  parties which was there is there and in future [11:45] there will be when you two human are talking  there is an element which cannot be diminished [11:52] by AI. So that judgment will be always there at  the end. We like to hire this person or no that [12:02] will be there. That's great. Thank you. And um  I love your company's tagline. So you'll maybe [12:09] mention that here. Uh but you as a consultant and  owner of an organization, you're talking to many [12:17] different companies. How are the expectations  of what they're asking you for changing? The [12:24] tagline you're referring to is the end of HR,  right? So definitely the end of HR as we know [12:30] in a way that we see more technical capabilities  as part of HR team these days across enterprises. [12:37] So there is an element of collaboration that is  more prominent than what we've seen before with [12:42] the chief HR people officer working incredibly  closely with the CTO's CIOS to try and understand [12:48] what's the right level of infrastructure that we  want to implement to deploy AI capabilities across [12:54] the full employee life cycle but there's also new  capabilities as part of the teams where we see [13:00] more engineering talent and technical people more  analysts they are kind of you know joining um the [13:06] HR function in a way that we've not seen before.  So there is this question around is the old way [13:11] of doing again HR something that's going to be  still relevant in the context of where the market [13:17] is heading and how companies are scaling and how  companies are evolving. do I need this many HR [13:23] business partner or can I scale in a different way  in a faster way but there's definitely a number of [13:28] assumption that's been challenged with kind of  a more technical lens also apply to the HR and [13:33] people teams absolutely because the scaling will  happen with people particularly um in a um session [13:42] I was moderating yesterday we were focusing on how  cocreation is so important as you scale because [13:49] you get the feedback you get the information  to move things forward. So, we've talked about [13:54] expectations and then now I'd like to focus on  the employee life cycle because that's what we [14:01] do. That's what we're offering, the employee  life cycle. We all know that the uh better the [14:08] employee experience is, the better the customer  experience is. So, why don't we start with [14:13] uh recruiting and Rosie um we're certainly here  talking about AI machines. How are you maintaining [14:21] a human element as part of the recruiting process?  Yeah, that's very good and sensitive question. [14:31] I know um we know in the industry right now when  these ATS system coming how much AI integration [14:40] is there but I believe as a leader in recruiting  a IML that we still needs to put the human in the [14:51] loop when we are looking at resume. I give you  this small example. If my leader is looking a [14:58] specific AI something and that is known that my  candidate has done that in their patent in their [15:08] publishing in PhD which is not there on their  two-page resume but when my recruiting team go [15:17] in and see there's a publication and they go on  Google Scholar or somewhere and see that yes this [15:24] person has done that so we are still missing that  using AI and we should not be like missing this [15:33] great talent that's still my point on human with  AI is still very important because there's also [15:41] like we can't have like six page of réumés right  we are like on two but we are missing a lot on [15:48] that thank you what about uh organizational  development how is it evolving well what [15:55] we're seeing at Pearson and from our enterprise  customers as well is that there's really a shift [16:00] from episodic development to really um continuous  capability building. And so I I there's absolutely [16:08] a place for the traditional kind of development  and onboarding. Um but I think what's important [16:15] right now is really to think about how do we help  people learn faster and so that they can learn as [16:21] fast as the work is changing. So that's really  what I spend a lot of my time thinking about. Um [16:26] and last year at Pearson, we launched um our new  Pearson power skills which are learning to learn [16:32] adaptability in AI for exactly this reason. We  need to create a workforce that is um adaptable, [16:39] can learn quickly, very proficient with AI. Um  and we think that this is a force multiplier. So, [16:46] um, we're hoping with this that our people can  build capability quickly, navigate change better, [16:53] um, and apply AI, uh, in a relevant and effective  way in their day-to-day work. Thank you. And so, [16:59] with all the services we have going on, how do  we make sure we're not disconnected from what the [17:05] business is doing and those initiatives? I I think  the key part which HR uh has its work cut out is [17:13] the organizational design and culture development  because what's happening right now I think uh with [17:21] a lot of changes you know now my organization  I I give you an example when covid came nobody [17:28] could predict that you know what would happen now  what we found was that when I mean because of our [17:36] strong sense sense of responsibility because  we are the uh biggest bank in Malaysia and [17:43] uh the sense of missiondriven and the sense of um  responsibility, social responsibility back to the [17:52] uh to the nation actually inspired people to uh  do things that were really innovative to solve [18:01] uh problems locally within the e economy.  For example, small, medium and enterprises [18:06] could not do business in a regular way. I mean,  customers were not coming during COVID. And so, [18:12] a lot of people, a lot of businesses would have  been demolished, right? So, what we did, I mean, [18:17] a group of people because they had that social  responsibility, you know, mindset and things [18:22] like that. They actually came together and  created a platform, an e-commerce platform [18:28] within 3 weeks. Within 3 weeks, it was launched uh  to market. Now being a bank of 60 year old like 60 [18:36] plus years developing anything within three weeks  and launching it to market is unheard of but what [18:43] made it work was no job description right because  this was covid times nobody could really predict [18:49] what's happening but what we found was the sense  of missiondrivenness the sense of purpose right [18:56] the sense of uh you know uh your service to your  customers that is what drives people uh uh and our [19:04] employees. And so when we talk about you know uh  rising technological changes and things like that, [19:10] we had a really good uh view that uh we really  need to inculcate the mission, the purpose, [19:17] the meaning into our jobs, right? And to not look  at jobs statically. Uh I I think we've I mean my [19:25] HRBPS and things like that we've been working  hard to change the mindsets of hiring managers [19:30] even right even ourselves that we shouldn't look  at typical job descriptions and not hire based on [19:35] task. No, we should be hiring based on values,  hiring based on, you know, enduring skills, [19:41] critical thinking, problem solving, you know, a  larger perspective, you know, interdisciplinary [19:46] perspective that people bring. And so that's how  we've been trying to change uh the culture as well [19:52] and and reinforce things that are enduring, right,  that can take us through different cycles. I love [19:57] that. And as we think about that, I'm I'm looking  at you now. if we could focus more on the ROI. So, [20:07] we're doing all of this work and we're working  with the leaders and they're investing, you know, [20:12] next year it's said to be $4 trillion um in  advanced technology, AI included. What are [20:22] they saying the leaders are working with about how  they get the return on those investments across [20:28] enterprise deployment? There are two key goals  that we that we see companies have and uh and [20:33] that they use to measure ROI. One is specifically  focus around operational efficiency um and the [20:39] other one is around performance impact and how  they scale there across every single individual [20:44] in the organization. Operational efficiency it  depends by the type of the companies but we see [20:49] um an angle around how do we make sure that we  can keep scaling the organization without having [20:55] a disproportionate amount of head as well.  So AI can be a multiplier to the impact of [21:02] individuals. So there is the identifying specific  areas where you can actually apply the multiplier [21:08] and invest more aggressively into those to get the  return that you would want. And in the context of [21:13] increased performance is around how do we remove  uh the work that we don't want or the interactions [21:19] that we don't want from incredibly high performing  individual in the company to make sure that they [21:22] can stay focused on driving again more impact  in their organization. Um another lens that [21:28] he's applied commonly is also the idea that agents  can be very similar to humans in the way they can [21:36] scale work but just the velocity at which they  scale is just much larger on specific type of [21:43] tasks. So there is an element apply which is also  in the same way you would think about headcount [21:48] scaling headcount you can scale agents as well  but with a fraction of the cost on specific type [21:53] of tasks. So staying with you a little bit,  what do the best companies do? Like give us [21:59] some of that secret sauce. I would say the best  companies are the one that incredibly intentional [22:05] about deploying AI capabilities. We seeing a  ton of companies that are experimenting with [22:11] pilots just to do some sort of you know checkbox  exercise. They want to say okay we are investing [22:16] in AI. We're thinking about AI. So we try and  do few things but the level of impact is not [22:22] as good as the company that say I want to pick up  something very very specific. Let's say I want to [22:28] automate the full onboarding process in a way that  is incredible from a user experience perspective, [22:33] incredible level of velocity, incredible level of  automation and they go incredibly deep with again [22:39] team composition technology to deploy and that's  where you see actually the value that is generated [22:45] where you actually are intentional and invest  time and effort into the change management piece [22:49] to make sure that the return is um is very much  there. Okay. Well, thank you. Now I'm I'm going [22:54] to stick with you for a second because uh you've  shared what the best are doing. You've talked [23:01] about that and and given us a sense for progress.  But where are organizations struggling? I'd say [23:10] where companies struggle is where they think too  much about the technology. Like we see a number of [23:15] conversation where people are thinking just about  AI we want AI but actually they're not thinking [23:21] what is the purpose of deploying AI and also they  underestimate the level of change management that [23:27] is required if you're not ready to make the change  if you're not ready to rethink again the the [23:33] incomposition the orchestructure who's going to  own AI what's the government's layer um delegation [23:39] of authority how is going to work it's just not  going to work that's the main thing that we see [23:44] where uh where companies think that they want AI  but actually they're not ready for the change more [23:49] than anything. Okay. Thank you. Well, let's stick  with this and uh Shuim, does that resonate with [23:56] you? Yeah, I think I I would agree there because  I think one of the greater challenges that I see [24:03] is the fact that um the technology piece is uh  you know it's a huge investment. So people are [24:13] very very cautious uh from a business side from  a technology pers functional perspective and [24:18] then from a people perspective because it's  it's creating a lot of changes and and this [24:24] is [laughter] and I saw one some of the surprises  I I saw when we were trying to digitalize the [24:31] organization was the push back from both the  tech side. Now I would I would think that the [24:36] tech people are you know onto this but they were  pushing back because they are responsible the [24:42] tech function is responsible to ensure uptime of  their re you know and to ensure that the systems [24:49] are delivering 100% so that they can there cannot  be any experimentations and failures. So that's on [24:55] one side then on the people side I mean people  are people and they they they want to continue [25:00] doing what they're doing you know everyone's busy  so when everyone's busy right trying to keep the [25:08] current state of affairs reliable and dependable  who creates the change so what we had to do was to [25:15] actually um you know create a task force now cross  function it has to be led crossf functionally so [25:22] create a separate you know cross functional team  but with people coming from within the businesses, [25:29] right? So, make it a a task force uh and it had  to be a special purpose project, right? Uh and [25:36] relook at the entire organizational design uh uh  you know and and we made it like a agile working [25:42] uh environment. So, so it has to be crossf  functionally let the the automation cannot [25:49] be just from a tech perspective. uh it has to be  from multiple perspectives you know because of the [25:56] ramifications uh of the changes if you want deep  changes in the organization share more about that [26:02] so you talked about crossf functional team say  more about who who led the team was it led out [26:10] of technology was it led from the business was  it led from HR because ultimately you do have to [26:17] have someone making those final decisions and then  what kind of rules of engagement did you all set [26:25] up to ensure that all of the voices were heard?  But first, let's talk about just the construct of [26:32] the team and who the leader was. Yeah. So, uh let  me share a bit of the the the background, right? [26:40] Uh now we know that to create behavioral  change uh one of the key levers that we had [26:48] to do was to look at the incentives for change  because if we don't change the incentives for [26:55] change I mean don't change incentives for  work people will continue working on the [27:00] same things. So incentives here does not mean  bonuses that is tied to digitalization. No, [27:08] it was uh non-monetary in the sense that one of  the things that we we looked at was uh uh giving [27:15] people promoting people who were champions of of  they were uh digitalization they wanted to create [27:23] changes. This therefore can be crossunctional. So  some of the pro uh some of the uh we had different [27:30] projects uh work streams. Some of the key work  streams were led by HR. Some of the key work [27:36] streams were led by business. Some of the key work  streams were led by tech. What ultimately whom [27:42] we chose was the leaders who typified the values  that we wanted in the organization, the leadership [27:49] values. And so that promoted that created the  incentive the the the because people wanted [27:55] to be part of this task force. They saw oh these  people were being recognized and therefore there [28:00] was a sense that I need to be part of this right.  Um we were projecting the kind of leaders that we [28:06] want you right multi-disciplinary uh tech you know  adoption led and change people really good people [28:14] managers who so we chose people not necessarily  for the technical expertise but for the leadership [28:19] expertise and who could talk crossf functionally  and who are great collaborators across so so that [28:25] was uh one way in which we also incentivize the  right things. Then of course we looked at the the [28:31] democ the tech platform. Now the one of the issues  that we had was that being a 60-year-old plus [28:37] organization a lot of things were in silos and so  uh we had to convince I mean there was a bit of [28:45] back and forth to convince tech to democratize  the tech stack right to democratize the data [28:52] because you can't run machine learning algorithms  with just a functional perspective. If you really [28:58] want good algorithms, the algorithms have to cut  across the organization to use variables right [29:04] from different parts of the organization. So the  infrastructure had to be democratized to be open [29:11] uh to different uh functional uh you know parts  of the organization. That's huge uh challenge [29:18] in itself because as a bank we have so many  regulations you know so many data protection [29:24] you know uh requirements and people are very very  uh uh um you know cautious on on who gets do you [29:30] see the data what kind of what kind of data and  and so forth. So the democratization of the infra [29:36] was another thing and the um tech and then uh of  course the work itself had to change accordingly. [29:44] So certain pieces, incentives, um, uh, tech, you  know, democratization, organization, design, these [29:51] were three key things that had to change. Uh, and  then res-killing could work because if these three [29:56] key things don't change, res-killing does not work  because you can train people, but if there's no [30:01] incentives for them to change and adopt the new  skills in the job, then it doesn't work as well. [30:08] Oh, that's great. And you also answered through  that response the rules of engagement and it [30:14] sounds like at the core of it were the values and  leading from that perspective but things happen [30:20] right Essen. So there are barriers to adaptability  and uh they can lead to not meeting expectations. [30:31] I mean we've all seen uh the the numbers and the  percentages 70% of of AI initiatives don't meet [30:39] their stated objectives etc. And it's not easy  getting the organization to adapt. So share your [30:47] thoughts on those barriers and how you all have  been able to overcome those. Sure. And my friends [30:52] up here actually have talked about quite a few of  them. So the one that I see as as quite common um [30:58] is when we treat AI as a tech deployment. And you  know we know that um the tech piece is easier to [31:04] deploy than the human piece. It's really hard to  get people to change the way they work and change [31:10] how decisions are made and really think about how  roles change and and and the value changes around [31:17] that. So I I would fully agree with my colleagues  here. Um, second thing, and I think we've touched [31:23] on this a bit as well, is when HR, tech, and  our business leaders are not in lock step. And [31:29] everyone's really excited about AI and they're all  driving their own change agendas with their own [31:35] priorities and timelines and commitments. Um, and  that doesn't work either because, you know, like [31:41] we said, it's it's not just a tech transformation,  it's a workforce transformation that requires [31:46] everyone to, you know, work together from the same  agenda. So that's again something we're uh trying [31:52] to drive at Pearson. Um and then the other thing  as well that um I'm fortunate that my my boss [32:00] does it. Um but one of the things that hinders  adoption is when our leaders don't role model the [32:04] right behaviors. Right? So, if they're out there  talking about AI in the organization, but they're [32:10] not learning about AI or they're not experimenting  or embedding it in their day-to-day activities, [32:16] um you know, it's it's hard to get people to  follow. Um and then a final point perhaps is [32:22] um and maybe this is more the the thing we  should do um is really if we want people to [32:29] adopt and learn about AI, we really need to embed  it in the rhythm of business as much as possible [32:34] um and make it a painkiller. Right? So um one  thing we've done at Pearson, I'm sure everyone [32:41] um loves the performance management process  in their organizations. That's something [32:45] I'm responsible for at Pearson. um and it's so  important but people are really busy delivering on [32:51] their business commitments and sales and so on. So  we have to make it as easy as possible for them. [32:56] So we've created a AI Asian uh called bar up which  helps our leaders, managers, individuals craft [33:04] really smart objectives and key results. Help them  reflect on their performance throughout the year. [33:10] help them craft their performance reviews so they  can focus on having really meaningful performance [33:16] conversations and get some help with the more  administrative side of it. So that's what we say [33:23] um that's one of the examples I use when I talk  about um AI as a painkiller. Wow. Thank you for [33:29] that. And I'm going to ask Rosie to to share. And  while she while she's getting ready to share, can [33:35] you all be thinking about leadership and what has  to change with the way leaders are leading today [33:45] to be effective in the AI era. So Rosie please  go ahead. Yeah, when we come to that point, see, [33:54] we all are looking for some kind of integration  of AI. But I believe in the organization, [34:02] we should think of learning first. Organization  we have how much learning we are doing with our [34:10] employees who are there are they adapting it. So  we can have any integration but the capacity of [34:18] that any integration is high but the usage how  much usage we have. So that's where it comes the [34:26] learning sessions a lot in the companies where  you can learn about AI adaptation. One example [34:34] is where AI is really helping recruiting now to  understand the depth of technology just like a [34:42] kindergartener which was not there before. So it's  very small but very impactful where now recruiter [34:51] sourcers can understand the depth of technology  what their hiring teams are hiring in their very [34:58] simple language and break it down however they  want to understand. So that is a power of AI I see [35:04] in the organization. So start where it's a usage  from the bottom to all the way up and leadership [35:11] team should look up at every stage of employment  how they are using what kind of learning we can [35:18] give them back um so that they can improve their  AI skills in the workforce. Thank you. So who [35:25] wants to start with the leadership question? Uh, I  think it's, and everyone's probably heard it, but [35:33] I think it's incredibly important for leaders to  demonstrate and role model learning and growing. [35:42] Uh for example, when my leaders were trying to  figure out what do we do with machine learning, [35:48] you know, back in, you know, back when machine  learning was a big thing, you know, 2018, 2017, [35:54] um one of the first things that they did was  to learn themselves. So our sea suite uh led by [36:01] the CEO and the board uh the first thing they did  was to take a two and a half day learning retreat [36:09] just to figure out what is this thing what is AI  what is an algorithm what is Python right how do [36:16] we actually code an algorithm right and so they  ro modeled in fact those days you know 20 impact [36:24] eight years ago right they actually learned  Python coding right and so we took pictures [36:29] of this and then we send it out to all employees  and say look if our leaders can take time to learn [36:36] time out and time to be vulnerable right time to  say I don't know and to shift the mindset from [36:43] a know it all to a learn it all learn right  learn and it's okay to fail uh it really uh [36:52] reduce the fears that people have the impostor  syndromes that exist this in many organizations, [37:00] right? Uh and to make it a more positive uh  learning and growing kind of institution and [37:08] that's the culture, right? Which leaders need  to role model more and more. You I mean this [37:15] has been going on for many years, right? But but  I think we really need to get it right because the [37:20] right answers can come from your interns, can  come from anyone now with the skills, right? [37:27] uh and I think we need to uh do that. Wonderful.  Thank you. It's definitely the culture. Essen, [37:35] thoughts from you. Trying to think of something  new here because uh I again I agree. I think it's [37:41] about how being open-minded, growth mindset, um  really embracing the learning culture. And uh at [37:48] Pearson as I mentioned earlier we have these uh  three power skills and we have proficiency levels [37:54] tied to each role. So our most senior people in  the organization actually are required to have [38:00] the highest proficiency level in AI. Now obviously  apart from engineers and so on people with very [38:06] technical roles. Um but the idea is things are  changing and they do need to be knowledgeable on [38:12] it and they do need to embrace it and role model  it. Um, and then I just add, um, more adaptable, [38:19] more resilient. Um, and we all need to move  faster because AI is really disrupting, um, [38:26] all of our markets, right? So, our leaders need to  be brave. They need to understand the technology [38:32] um, and how it's disrupting their businesses and  and get ahead of it. Okay. Thank you, Javon. Um [38:38] we talk a length about change and learnings and on  top of that what we had is the leaders need to be [38:42] incredibly comfortable today with uh ambiguity and  navigating ambiguity as well. Um leader tends also [38:49] to trust appear massively but AI is incredibly new  like you know two years ago was not even a thing [38:56] effectively or in the same in the same way that is  today and you don't have enough social proof and [39:02] validation to kind of you know have conversation  and ask people tell me exactly what you're doing [39:06] the impact the result and so on. So everybody's  learning everybody's trying to kind of you know [39:11] figure this out. So you need to be incredibly  comfort with uh with some level of discomfort, [39:16] let's say, to move fast and and break a few  things, but hopefully for for good reasons. [39:21] Thank you. And uh Rosie, anything to add there?  Yeah, for leadership spend more time with the [39:30] team down to understand what they're doing. I  think they will learn more from their team now. [39:38] the interns the young engineers who are coming in  and see that um I think it's a teamwork at this [39:47] point with the leadership not in the silo just  leaders are there because if you see we all are [39:53] generation by generation and we need to come down  together to learn this AI and human together some [40:01] pieces believe in like human touch more somebody  is saying AI so we need to collaboratively work [40:08] So I would say leadership needs to collaborate  more with their teams. Okay. So staying with you, [40:14] the organization that thrives in the  AI era will be the ones that what? So [40:23] the organizations that thrive in the AI era,  finish this sentence, will be the ones that [40:32] learn from each other. There's a human still  human. So we all are going at different resources [40:39] different time we are not robot so human come  together so that's where I personally feel very [40:47] strongly human touch excellent thank you Giovani  we move fast and embrace change okay go ahead [40:55] um build human capability with the same  intentionality that they deploy technology [41:03] um convert artificial intelligence to  enlighten intelligence. And I'm going to [41:11] borrow that phrase enlightened intelligence  from one of the researchers from my [41:17] uh Indiana University who I met yesterday um  who really is talking about uh you know make [41:24] sure that humans continue to cultivate  the sense of meaning purpose and values [41:31] right to do this responsibly um for human  driving. Wonderful. Very last question. [41:40] Let's leave the audience with this. Uh what's  one capability? Uh oh, my thank you. What is [41:50] one capability that organizations should  begin building right now for the AI era? [42:01] You want to start? Oh, okay. Um the I think the  ability to you know figure out how do you best [42:13] um move humans up the value chain. I think today's  keynote address spoke about that. How do we truly [42:22] augment human capability and I think preserve  that meaning and create that shared prosperity. [42:31] I think that is what organizations should do.  How do we create a bigger pie for everyone to [42:37] thrive? I think that I would continue to echo  today's keynote address. Isn't that wonderful? [42:43] Go ahead. I would say organizational learning  agility and I think that's where competitive [42:50] advantage is going to come from. organizations  who can look ahead, look around corners, [42:55] understand AI and really understand the  capabilities needed in their workforce most [43:00] quickly build them, most quickly apply them  so they can actually create business results, [43:05] value and impact. Those are going to be the ones  with the competitive advantage. I would say also [43:12] identify the areas where AI can drive impact and  then double down on people's skills that can't [43:19] be necessarily can be translated and uh and yeah  just um focus focus on people's skills. Rosie. Oh, [43:28] I would say governance piece will be a big one  which we should have governance. We should not [43:36] ignore that. Absolutely. So we are wrapping up and  I'd say just across all facets of HR from culture, [43:47] talent, leadership and organization, we have to be  completely tied with our partners from the CEO to [43:58] the finance leaders to the technology leaders, the  business leaders for us to truly affect change. [44:04] building a coalition together to drive  the organization forward very fast is [44:11] super critical and uh I appreciate all of the  insights from you all today and helping us to [44:17] get there and thank you to the audience for  joining us. We appreciate it. Thank you all.