Oklahoma Outreach Lifts AI Use to 74% by Tackling Fear Over Tech

Panel 5 - 3: Leading Human-Centered AI Adoption: From Awareness to the Human Edge — Ai4

Ai4

Enterprise AI adoption stalls from unaddressed human emotion rather than software limits, and treating implementation as adult learning rather than an IT rollout is what unlocks workplace fluency.

AI-assisted summary of Panel 5 - 3: Leading Human-Centered AI Adoption: From Awareness to the Human Edge — original source. Check the original for full context.

Source published:

Detailed notes

## Background Published on YouTube by Ai4 on August 14, 2026, this presentation features Belinda Biscoe, Senior Associate Vice President for University Outreach at the University of Oklahoma. The session outlines a three-year, data-backed initiative that guided an organization of 1,400 employees from AI skepticism and fear to confident fluency by prioritizing human behavior over technology.

## TL;DR Belinda Biscoe details a longitudinal case study showing how the University of Oklahoma's College of Continuing Education overcame intense workplace anxiety surrounding generative AI. Utilizing a five-stage framework rooted in psychological safety, empirical sentiment analysis, and continuous adult learning, the division achieved a 74% adoption rate within one year without executive mandates. As training progressed, initial dread evolved into mature ethical inquiry, with 95% of employees ultimately articulating their irreplaceable human edge. Biscoe emphasizes that successful digital transformation depends on treating AI adoption as an adult education challenge rather than a standard IT rollout.

## Atmosphere & style The presentation carries an authoritative yet deeply empathetic tone, anchored by Biscoe's dual perspective as an organizational leader and research psychologist. She connects with the audience through candid, relatable anecdotes of early workforce resistance before systematically walking through data-backed outcomes. The delivery feels pragmatic, reflective, and empowering, balancing rigorous organizational research with an inspiring reaffirmation of human judgment.

## Key takeaways - [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) The primary barrier to successful enterprise AI adoption is human emotion and behavior rather than software capability. - [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) Investing in people before technology helped a 1,400-person workforce transition from 50% having zero AI experience to 95% articulating their irreplaceable human value. - [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) Baseline organizational resistance can manifest as extreme fear or moral objection, making empirical sentiment tracking and psychological research essential. - [04:53](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=293s) Employee fear should be treated as a legitimate baseline and mandatory starting line rather than an operational failure to be dismissed with cheerleading. - [04:53](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=293s) Leaders must visibly participate as fellow learners to model curiosity and build psychological safety across the organization. - [07:48](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=468s) Structured, non-mandatory adult learning drove AI usage from 50% zero-experience to 74% regular or occasional adoption within a single year. - [07:48](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=468s) Employee skepticism regarding algorithmic bias and output fairness reflects critical thinking and ethical maturity rather than negative resistance. - [07:48](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=468s) 89% of employees affirmed that essential human skills—including empathy, contextual judgment, and ethical decision-making—became more strategic in an AI-embedded workplace. - [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) Pre-assessment sentiment discovery (Stage 2) and the 'Human Edge' (Stage 5) are the two critical phases organizational leaders most commonly skip during AI rollouts. - [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) AI adoption must be framed and managed as an adult learning process rather than an IT deployment to avoid alienating non-technical employees. - [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) Professional development must deliver role-tailored practical workflows rather than generic, high-level software feature demonstrations. - [15:36](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=936s) Long-term competitive advantage in the AI era will belong to organizations that prioritize the human side of transformation over tool procurement.

## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) Introduction and the Journey to the Human Edge

Belinda Biscoe introduces herself as the Senior Associate Vice President for University Outreach in the College of Continuing Education at the University of Oklahoma. As head of one of the nation's largest continuing education organizations, she oversees approximately 1,400 employees across 17 distinct departments. Addressing an audience spanning academia, corporations, nonprofits, and industry, Biscoe presents a three-year case study titled *"Leading Human-Centered AI Adoption: From Awareness to the Human Edge."* Her central thesis challenges the tech-first narrative dominating enterprise adoption: the primary barrier to digital transformation is not software capability, but human emotion and behavior.

Reflecting on the baseline conditions in 2023–2024, Biscoe reveals that half of her workforce had never used or touched an artificial intelligence tool. Reactions ranged from outright fury to paralyzing fear, with multiple staff members convinced their professional careers had reached an abrupt end. By 2026, after intentionally investing in people before tools, 95% of those employees could articulate precisely what makes them irreplaceable in an AI-embedded workplace. Rather than pushing rapid rollouts or top-down mandates, the organization unlocked fluency by directly addressing emotional paralysis through a structured, people-first roadmap.

- **Workforce Size:** Approximately 1,400 employees (800 full-time staff, 600 part-time and occasional workers). - **Core Shift:** From 50% having zero AI experience in 2023–2024 to 95% having clear conviction of their unique, irreplaceable human value by 2026.

---

[02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) Gauging Resistance, Early Backlash, and Mixed-Methods Research

Engaging the room, Biscoe asks how many organizational leaders still grapple with employees terrified of AI, or have invested in expensive licenses only to watch tools sit unused. Hands shoot up across the audience. Identifying herself as a baby boomer who nonetheless tends toward early technology adoption, Biscoe notes that she began experimenting with generative AI roughly five years prior. Recognizing immediately that the technology would be a profound game-changer, she resolved to build comprehensive AI literacy across the entire college.

Initial resistance was extreme. In a baseline, college-wide survey distributed to assess organizational readiness, one employee responded with the verbatim assessment: "This is against God and humanity." Others openly declared they would quit their jobs if the college forced AI implementation. To navigate this friction systematically, Biscoe leaned into her background as a research psychologist. She established a rigorous mixed-methods study spanning three years across all 17 departments, combining qualitative sentiment analysis with quantitative surveying before and after every training initiative.

- **Baseline Sentiment:** 28% of employees in Year 1 believed AI offered zero meaningful benefits to their daily work. - **Methodology:** Longitudinal mixed-methods design pairing broad annual surveys with pre- and post-session evaluations for all professional development modules.

---

[04:53](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=293s) Creating Psychological Safety and Continuous Professional Development

Rather than ignoring employee apprehension or brushing it aside with corporate cheerleading, leadership made a calculated choice to treat fear as a legitimate baseline. Biscoe emphasizes that leadership deliberately avoided sending hollow, celebratory emails declaring, "Oh, this is so exciting." Instead, they established psychological safety by acknowledging that all major technological disruptions introduce genuine challenges alongside operational advantages. In her view, employee fear is never an operational failure—it is simply the mandatory starting line.

The centerpiece for organizational learning became the annual "State of Outreach" conference, a large-scale gathering repurposed to deliver deep professional development. Rather than treating training as an isolated, one-and-done event, the college structured continuous, multi-tiered learning emphasizing organizational learning, workforce readiness, ethical reflection, and practical hands-on application. Survey data gathered prior to the programming directly dictated the curriculum. In later phases, this included voluntary deep-dive training on Microsoft Copilot, ensuring that instruction matched staff readiness rather than administrative assumptions. Throughout, Biscoe visibly participated as a fellow learner, reinforcing curiosity from the top down.

```mermaid flowchart TD accTitle: The Five-Stage Human-Centered AI Framework subgraph Foundation ["1. Alignment & Baseline"] A["Stage 1: Leadership Commitment<br/>& Demonstrated Curiosity"] --> B["Stage 2: Pre-Assessment & Sentiment Discovery<br/>(Fears, Needs, Literacy)"] end subgraph Implementation ["2. Literacy & Context"] B --> C["Stage 3: Organization-Wide Shared Awareness<br/>(Common Vocabulary & Safety)"] C --> D["Stage 4: Role-Specific Training<br/>(HR, Finance, Research, Admin)"] end subgraph Maturation ["3. Strategic Edge"] D --> E["Stage 5: The Human Edge<br/>(Critical Inquiry, Ethics & Irreplaceable Skills)"] end ```

---

[07:48](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=468s) From Existential Dread to Operational Sophistication

The impact of this approach materialized rapidly. Within a single year—without mandating usage, procuring a disruptive new AI platform, or exerting coercion—the college's metrics inverted: zero-experience rates collapsed from 50%, while 74% of the workforce reported using AI regularly or occasionally. Crucially, employee concerns did not vanish; instead, they matured. What began as existential dread about employment survival transformed into operational sophistication and critical inquiry.

Staff members across all administrative functions—educators, HR specialists, grant writers, and frontline customer service staff—began evaluating algorithmic bias and actively questioning automated outputs. Employees moved toward asking nuanced, systemic questions, such as: "Who benefits more from an AI-automated process?" For Biscoe, skepticism shifted from being a symptom of obstinate resistance into concrete proof of organizational growth, ethical leadership thinking, and critical maturity.

- **Adoption Surge:** 74% of staff using AI regularly or occasionally after 12 months of structured learning. - **Role Value Recognition:** 89% of employees affirmed that their human roles had become *more* critical and strategic in an AI-embedded workflow. - **Key Human Assets Identified:** Ethical decision-making, contextual judgment, empathy, and cross-functional collaboration.

---

[11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) The Five-Stage Framework and Key Leadership Lessons

Biscoe formalizes the college's methodology into a transferable five-stage model, highlighting Stages 2 and 5 as the steps leaders most frequently skip when rolling out new systems. Stage 1 requires leadership to make an explicit commitment and model vulnerability as active learners. Stage 2 mandates continuous empirical surveying to pinpoint what employees fear, know, and require before designing any learning intervention. Stage 3 builds organization-wide awareness and shared vocabulary, unifying staff through common experiences. Stage 4 delivers role-specific application tailored to functional duties—such as budgeting for accountants, talent management for HR, or literature synthesis for grant writers—honoring adult learning theory. Stage 5 targets the "Human Edge," elevating irreplaceable cognitive and interpersonal assets.

From this framework, Biscoe extracts fundamental operating principles: AI adoption is an adult learning challenge, never a standard IT software rollout. Framing implementation as an IT project alienates non-technical workforces immediately. Furthermore, so-called "soft skills" must be recognized as strategic assets that serve as an organization's primary competitive advantage in an automated landscape.

- **Actionable Recommendations for Leaders:** - Survey comprehensively *before* configuring training curricula. - Establish baseline literacy and common vocabulary before tool deployment. - Create collective, low-stakes shared experiences across departments. - Deliver role-tailored instruction rather than generic feature demos. - Build explicit ethical frameworks alongside technical workflows. - Treat human skills as hard strategic assets rather than secondary soft skills.

---

[15:36](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=936s) Open Questions, Study Limitations, and Call for Collaboration

Concluding her presentation, Biscoe outlines the institutional boundaries of the initiative and poses critical research questions that remain unanswered. Because the study was conducted within a single university continuing education division, the findings are compelling but not yet universally generalizable across corporations, government agencies, or alternative nonprofit ecosystems. Additionally, survey participation was voluntary, introducing potential self-selection bias among more engaged staff, while survey instruments evolved from year to year rather than remaining static.

Biscoe leaves the audience with core questions currently driving her academic inquiries: how sustained AI adoption affects long-term employee well-being and productivity, and precisely which human capabilities correlate most strongly with resilient AI integration. She invites attendees to partner on comparative, multi-site studies and longitudinal tracking across other enterprise environments, reiterating her overarching philosophy: "The organizations that will win with AI in this era will not be the ones with the best tools; they will be the ones who figured out the human side first."

## Notable quotes > I've watched an organization of 1,400 people go from fear to fluency with AI, and I can show you exactly how we did it. — Belinda Biscoe

> Whether you lead a Fortune 500 team, a startup, or a university, the #1 reason AI adoption stalls isn't the technology; it's the people, and most leaders are skipping the steps that actually change behavior. — Belinda Biscoe

> the most powerful outcome of our AI initiative wasn't that people used AI more; it's that they became clearer about what makes them irreplaceable. — Belinda Biscoe

> [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) Leading Human-Centered AI Adoption: From Awareness to the Human Edge. — Belinda Biscoe

> [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) This is against God and humanity. — Survey Respondent

> [04:53](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=293s) Oh, this is so exciting. — Belinda Biscoe

> [07:48](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=468s) Who benefits more from an AI-automated process? — Staff Member

> [15:36](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=936s) The organizations that will win with AI in this era will not be the ones with the best tools; they will be the ones who figured out the human side first. — Belinda Biscoe

## Mentioned Belinda Biscoe, University of Oklahoma, College of Continuing Education, University Outreach, Ai4, Microsoft Copilot, State of Outreach Conference

## Video description Belinda Biscoe, Senior Associate Vice President for Outreach/College of Continuing Education, The University of Oklahoma

I've watched an organization of 1,400 people go from fear to fluency with AI, and I can show you exactly how we did it. Whether you lead a Fortune 500 team, a startup, or a university, the #1 reason AI adoption stalls isn't the technology; it's the people, and most leaders are skipping the steps that actually change behavior. In this session, I'll share three years of real data, a five-stage framework you can take back to your organization Monday morning, and the counterintuitive truth we discovered: the most powerful outcome of our AI initiative wasn't that people used AI more; it's that they became clearer about what makes them irreplaceable. Come ready to rethink your AI strategy, because the organizations that will win aren't the ones with the best tools; they're the ones who figured out the human side first.

[Source](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk)

中文简报

化解技术恐惧：俄克拉荷马大学将AI采用率提升至74% 企业在AI采用上的停滞源于未被化解的人类情绪，而非软件本身的局限；将实施过程视作成人学习而非IT部署，才是释放职场应用能力的关键。 Belinda Biscoe 详细介绍了一项纵向案例研究，展示了俄克拉荷马大学继续教育学院如何克服职场中对生成式 AI 的强烈焦虑。该部门依托植根于心理安全感、实证情绪分析和持续成人学习的五阶段框架，在无需高管行政命令的情况下，于一年内实现了 74% 的采用率。随着培训的推进，最初的恐惧逐渐转变为成熟的伦理层面的探讨，最终有 95% 的员工能够明确阐述自身不可替代的人类独特优势。Biscoe 强调，成功的数字化转型取决于将 AI 采用视为一项成人教育挑战，而非单纯的标准 IT 部署。 企业推进 AI 落地的核心阻碍在于人的情绪与行为而非技术本身；俄克拉荷马大学通过优先投资于人，使 1,400 人的团队在三年内从半数零经验转变为 95% 的员工能清晰阐述自身不可替代的独特价值。 (0:04) 面对初期的强烈恐惧与道德抵触，管理层摒弃了空洞的说教，将员工担忧确立为正当的基准起点，并通过为期三年的混合方法学实证研究持续跟踪全员心理与需求。 (2:16) 通过推行非强制、分阶段的成人学习机制，该组织在一年内将 AI 常规或偶尔使用率大幅提升至 74%，员工的顾虑也从生存恐慌逐步升华为对算法偏见和公平性的深度批判性思考。 (7:48) 89% 的员工确认同理心、情境判断与伦理决策等能力在 AI 深度融入的工作流中更具战略意义，表明 AI 转型本质上是成人学习与人本优势的升级，而非单纯的 IT 软件部署。 (11:49)

Reading summary

Belinda Biscoe, Senior Associate Vice President for University Outreach at the University of Oklahoma, argues that the fundamental barrier to enterprise artificial intelligence adoption is human emotion and behavior rather than technological capability. Drawing on a three-year longitudinal study of 1,400 employees across 17 departments, Biscoe asserts that organizations achieve sustainable adoption only by establishing psychological safety, empirical sentiment discovery, and adult learning frameworks before deploying software. Addressing Emotional Resistance Through Longitudinal Research (2:16) Assessing baseline conditions in 2023–2024, the university division found that 50% of its workforce had never used an artificial intelligence tool, and 28% believed AI offered zero meaningful benefits to their daily work. Staff reactions ranged from anxiety and professional grief to outright hostility, with some employees threatening resignation and one survey respondent declaring the technology was "against God and humanity." Rather than ignoring apprehension or offering superficial corporate encouragement, leadership treated fear as a predictable and valid starting line. Drawing on her background as a research psychologist, Biscoe instituted a three-year mixed-methods study combining annual college-wide surveys with pre- and post-session evaluations for every professional development module. By measuring sentiment, fear, and literacy before introducing specific systems, the college avoided premature technology procurements and tailored learning pathways directly to empirical workforce needs rather than executive assumptions. Psychological Safety and Role-Tailored Adult Learning (4:53) Rather than relying on coercive top-down mandates or treating instruction as a one-time event, the division embedded continuous professional development into recurring forums like the annual "State of Outreach" conference. Curricula emphasized workforce readiness, ethical reflection, and voluntary hands-on modules, including optional deep-dive sessions on Microsoft Copilot. Biscoe actively participated as a visible fellow learner during training, signaling executive vulnerability and reinforcing curiosity from the top down. The educational strategy was grounded in adult learning theory rather than generic software feature overviews. Instruction was customized to the concrete workflows of distinct operational units, ranging from grant literature synthesis to HR talent management and financial budgeting. Biscoe emphasizes that framing implementation as an adult learning challenge prevents the immediate workforce alienation that typically results when AI integration is treated as an IT-led software rollout. Transforming Existential Dread into Critical Operational Maturity (7:48) Within twelve months of structured, non-mandatory instruction, the college's metrics inverted: zero-experience rates collapsed, and 74% of the workforce reported using AI tools regularly or occasionally. Employee concerns did not disappear, but they matured from existential dread about job loss into operational sophistication and critical inquiry. Staff across administrative, instructional, and frontline roles began actively auditing outputs, evaluating algorithmic bias, and questioning who benefits most from automated processes. By 2026, 95% of employees reported a clear understanding of what makes them irreplaceable in an automated workplace, up from the baseline where half had no tool familiarity. Furthermore, 89% affirmed that their human roles had grown more critical and strategic alongside AI integration. The organization identified context judgment, ethical evaluation, empathy, and cross-functional collaboration as irreplaceable human capabilities that automation cannot replicate. The Five-Stage Framework for Human-Centered Implementation (11:49) Biscoe formalized the division's methodology into a five-stage model: leadership commitment and curiosity (Stage 1), pre-assessment and sentiment discovery (Stage 2), organization-wide awareness and shared vocabulary (Stage 3), role-specific training (Stage 4), and cultivating the "Human Edge" (Stage 5). Biscoe notes that enterprise initiatives stumble most frequently because leadership bypasses Stage 2's pre-assessment surveys and Stage 5's emphasis on non-technical cognitive competencies. The framework repositions interpersonal and ethical capabilities—often dismissed as soft skills—as an organization's primary competitive advantage in an automated landscape. Actionable recommendations include conducting comprehensive surveys before designing curricula, establishing a common vocabulary to demystify tools, delivering role-tailored instruction instead of generic vendor demos, and developing explicit ethical guidelines alongside technical workflows. Biscoe highlights several study limitations: the research took place within a single university continuing education division, participation was voluntary (introducing potential self-selection bias among more engaged staff), and survey instruments evolved over the three-year period. Unresolved questions remain regarding how sustained AI adoption affects long-term employee well-being and productivity, and which specific human capabilities correlate most strongly with resilient technical integration across broader corporate, government, and nonprofit sectors.

精炼阅读摘要

俄克拉荷马大学继续教育学院高级副协理副校长 Belinda Biscoe 提出，企业应用人工智能的核心障碍并非软件技术本身，而是员工的情感与行为；通过建立正视员工恐慌的心理安全机制与以成人学习为核心的五阶段框架，组织能够在不强制推行技术的前提下，将抵触情绪转化为不可替代的“人类优势”。 将恐惧视为起点：重塑组织心理安全 (2:16) 在2023至2024年的基线调查中，俄克拉荷马大学继续教育学院（University Outreach in the College of Continuing Education）约1,400名员工（包含800名全职及600名兼职与临时人员）中，有一半人此前从未接触或使用过人工智能工具。面对新技术引入，员工情绪从强烈的愤怒蔓延至职业生存恐慌，更有受访者在全院调查中直言“这违背了上帝与人性”，甚至有人以辞职相威胁；在第一年，有28%的员工认为 AI 对日常工作没有任何实际助益。 面对这种极端抵触，Biscoe 指出领导层并未采取自上而下的行政命令，也摒弃了轻描淡写或虚饰太平的“企业动员话术”。管理层明确将员工的恐慌视作数字化转型的合理基线，公开承认颠覆性技术必然伴随现实困境与阵痛。学院将“恐惧并非运营失败，而是必经的起点”作为共识，通过承认技术冲突构建了基本的心理安全感，避免了工具采购后被员工闲置的局面。 以成人学习为导向的五阶段框架 (11:49) Biscoe 强调，AI 采纳本质上是一场成人学习挑战，若将其矮化为单纯的 IT 软件部署，往往会直接疏离非技术岗位的基层员工。为此，学院推行了为期三年的混合方法研究，在17个部门中结合定性情绪追踪与培训前后的定量评估，将每年的“外展现状”（State of Outreach）大会改造成分层进阶的学习平台，并依据调研反馈动态调整教学内容，包括后续针对 Microsoft Copilot 开展的自愿深入培训，管理层自身亦作为共同学习者全程参与。 学院在此基础上提炼出标准化的五阶段人本模型：第一阶段要求领导层确立承诺并展现求知欲；第二阶段开展针对恐惧、需求与数字素养的情绪调研与前置评估；第三阶段构建全院统一的语汇与无压力的通识认知；第四阶段根据成人学习特征，提供面向人力资源、财务或项目申请等岗位的角色定制培训；第五阶段则进阶至“人类优势”（The Human Edge），聚焦算法无法替代的思辨与人际能力。Biscoe 观察到，第二阶段的情绪前置调研与第五阶段的优势挖掘，正是多数组织在推进技术采纳时最常忽视的环节。 认知跃迁：从生存恐慌走向操作成熟 (7:48) 在未强制要求工具使用、未采购颠覆性新平台的情形下，学院仅通过一年的系统性赋能便实现了指标逆转：完全缺乏经验的员工比例大幅下降，74%的受访员工表示已能够定期或偶尔在工作中使用 AI。更为深层的变化在于员工提问视角的成熟——最初的失业恐慌演变为了业务层面的深度审视与批判性思维。 来自教学、人力、项目申请撰写及一线客服等职能的员工，开始自发识别算法偏见并检验自动化生成内容的可靠性，进而探讨“在 AI 自动化流程中，究竟谁是更大的受益者”等伦理议题。调查显示，89%的员工认同人类角色在嵌入 AI 的工作流程中变得更加关键与战略化；到2026年，95%的员工能够清晰阐明自身在人机协同工作环境中的独特性与不可替代性。Biscoe 认为，同理心、情境判断、伦理决策与跨部门协作等传统“软技能”，在此过程中已被证实为组织应对自动化变革的核心战略资产。 Biscoe 指出，该研究局限在于数据仅来自单一高等院校的继续教育部门，其成果向私营企业或政府机构推广时仍待检验；同时，调查依托自愿参与可能带来样本自选偏差，问卷测量工具在三年中亦有迭代。针对 AI 持续采纳对员工长期福祉与生产率的纵向影响，以及何种人类能力最能促成韧性整合，Biscoe 呼吁各界开展多中心对比研究，并重申其核心观点：在 AI 时代胜出的组织，靠的不是拥有顶尖工具，而是率先攻克了人本转型议题。

Detailed notes — zh

## Background 该内容由频道 Ai4 于 2026 年 8 月 14 日发布在 YouTube 平台上。页面简介显示，俄克拉荷马大学外展与继续教育学院高级副校长 Belinda Biscoe 在本场演讲中分享了其带领 1,400 人团队从恐惧走向熟练运用 AI 的三年真实历程，并介绍了一套可立即落地的五阶段人本 AI 应用框架。 ## TL;DR 面对最初团队中半数员工零 AI 经验且伴随极度恐慌甚至扬言辞职的严峻阻力，Biscoe 并没有强行推行技术采购，而是将重心放在为员工构建心理安全感和开展定制化职业发展培训上。经过三年的循序渐进推进，学院不仅实现了 74% 以上的日常使用率，更有高达 95% 的员工明确掌握了自身在算法时代独特的不可替代价值。Biscoe 提出了涵盖领导力示范、事前调研、全员共识、岗位深耕和确立人类优势的五阶段框架，强调 AI 落地的关键在于成人学习与人类行为转变，批判性思维与伦理判断力等软技能才是企业最核心的战略资产。 ## Atmosphere & style 整场分享散发着学者型领导者的严谨与温情，语调真诚、笃定且极具感染力。演讲从最初员工甚至认为技术“违背神明”的剧烈阻力切入，展开了一段从集体焦虑到理性觉醒的动人叙事弧线。聆听过程既像是在观摩一份扎实的组织变革案例研究，又像是在接受一场富有同理心的领导力公开课，平实的数据与坦诚的科研局限性反思交织，令人感到踏实而振奋。 ## Key takeaways - [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) AI 落地受阻的首要原因从来不是技术本身，而是组织忽视了人类心理与行为转变规律；优先投资于“人”才是推动变革的根本路径。 - [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) 俄克拉荷马大学继续教育学院在三年内实现变革，零 AI 经验员工从 50% 降低，全院 95% 的员工不仅能熟练应用 AI，且明确了自己的不可替代性。 - [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) 推行初期面临严峻的抵触情绪，28% 员工认为 AI 毫无用处，甚至有员工以辞职抗议；领导层必须正面正视恐惧并营造心理安全感，而非通过通稿或命令强推。 - [06:50](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=410s) AI 培训不可采用一刀切模式，必须坚持“先调研后设计”，根据各部门与岗位的真实需求定制包含素养、伦理与实操的职业发展项目。 - [06:50](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=410s) 在无强制指令与采购新平台的前提下，通过针对性培训，学院定期或偶尔使用 AI 的员工比例在一年内由 50% 快速跃升至 74%。 - [06:50](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=410s) 员工关切点从“对生存危机的恐慌”转化为“对业务精细度的探讨与算法伦理的质疑”，体现了组织向更高成熟度迈进。 - [10:21](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=621s) 89% 的员工认为在 AI 深度嵌套的世界中自身岗位价值不降反升，批判性判断力、伦理决策与复杂协同等“人类独特优势”成为核心壁垒。 - [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) 总结出五阶段人本 AI 应用框架：涵盖领导示范、需求研判、建立通用素养与统一词汇、特定岗位定制培训，以及最终确立人类边缘优势。 - [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) AI 应用本质上是成人学习与行为改变项目，一旦将其错误归类为纯 IT 项目，组织就会失去大众的广泛参与和支持。 - [16:00](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=960s) 该研究目前基于单一机构且存在自愿参与的样本局限，团队正推进纵向跟踪体系并寻求跨行业合作，以验证不同环境下的普适性与长期影响。 ## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) 重新定义 AI 落地：从全员恐慌到确立“人类独特优势” 俄克拉荷马大学（University of Oklahoma）外展与继续教育学院高级副副校长 Belinda Biscoe 登台分享了她带领庞大团队应对 AI 变革的真实案例。作为全美规模最大的继续教育机构之一，该学院拥有约 1,400 名员工（包括 800 名全职人员及 600 名兼职/临时员工），下辖 17 个部门。在 2023 至 2024 年间，团队中约有一半的人从未接触或使用过 AI。面对这项未知的新技术，员工的反应极其剧烈：有人感到极度恐慌、愤慨，甚至坚信自己的职业生涯即将画上句号。 面对团队的集体焦虑，Biscoe 并没有选择直接采购最新的技术工具，而是将突破口放在了“人”的身上。经过三年的持续推动，到了 2026 年，学院高达 95% 的员工不仅能够熟练运用 AI，还能清晰准确地阐述在 AI 深度嵌套的世界中，究竟是什么让他们自己“不可替代”。Biscoe 指出，AI 落地停滞不前的首要原因从来不是技术本身，而是组织忽视了人类心理与行为转变的规律。 - **组织规模与结构**：涵盖 1,400 名员工（800 名全职、600 名兼职），分布于 17 个不同部门。 - **态度转变对比**：2023–2024 年有 50% 员工零 AI 经验且伴随强烈恐惧；到 2026 年，95% 的员工明确掌握了自己的不可替代性。 - **核心理念**：技术采购无法直接带来应用，优先投资于“人”才是推动变革的根本路径。 --- ### [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) 破除恐惧与阻力：从“违背神明”到确立心理安全感 作为一名自采婴儿潮一代（Baby Boomer）但属于早期技术采用者的领导者，Biscoe 在 5 年前便开始探索 AI，并深刻意识到这将彻底改变游戏规则。为了在学院普及 AI 素养，她开展了一项全院问卷调查，而收集到的反馈令人震惊——有员工在问卷中直言不讳地写道：“这项技术违背了上帝与人性（This is against God and humanity）。”甚至有多名员工明确表示，如果学院强制推进 AI 应用，他们宁愿选择辞职。 在第一年的调查中，至少有 28% 的员工认为 AI 没有任何实质益处。面对如此严峻的抵触情绪，Biscoe 强调领导层绝不能轻视或逃避这种恐惧，更不能只发一封“AI 非常令人兴奋”的通稿邮件了事。相反，领导层必须拥抱问题，为员工营造充分的“心理安全感（Psychological safety）”。她指出：“恐惧并非失败，而是变革的起点。” ```mermaid flowchart LR accTitle: AI Adoption Evolution Flow subgraph S1["心理阻力期"] A["存在感恐惧/抵触"] --> B["心理安全感建立"] end subgraph S2["能力提升期"] B --> C["基础素养与技能"] end subgraph S3["价值重塑期"] C --> D["业务融合与人类优势"] end ``` - **真实抵触数据**：首年调查显示，28% 的员工坚信 AI 没有任何实际价值，部分员工甚至以辞职相要挟。 - **领导力应对策略**：不回避问题、不强制推行，通过承认新技术的利弊来构建心理安全感。 - **科研方法论**：采用混合方法研究（Mixed-methods study），在三年间持续进行问卷调查与职业发展培训（Professional Development）。 --- ### [06:50](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=410s) 持续性职业发展与能力跃迁 为了将理念转化为行动，Biscoe 团队利用每年举办的“外展州况大会（State of Outreach）”作为契机，推出了密集且长期的职业发展培训项目。这种培训并非一刀切的单场讲座，而是涵盖了组织学习、劳动力准备度、伦理反思以及实际应用。在后续年份中，团队还面向自愿报名参加的员工提供了深度的 Microsoft Copilot 专项实操培训。 更为关键的是，所有的培训课程与内容设计，都是在收集并分析了培训前的调查问卷后才定稿的，真正实现了“按需施教”。Biscoe 自身也以身作则，向全院展现出好奇心与“终身学习者”的姿态。仅经过一年的专项投入，在没有强制指令、没有采购新 AI 平台、没有任何胁迫的情况下，零 AI 经验的人数比例大幅下降，定期或偶尔使用 AI 的员工比例从 50% 快速跃升至 74%。 - **应用率飞跃**：仅一年时间，定期/偶尔使用 AI 的员工比例从约 50% 飙升至 74%。 - **关切点的演变**：员工的顾虑并未消失，而是从“对生存危机的恐慌”转化为“对业务操作精细度的探讨”，这被视为组织成熟的标志。 - **覆盖岗位类型**：包括行政人员、教育工作者、人力资源专业人员、客户服务代表、资助申请撰写人（Grant Writers）等。 --- ### [10:21](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=621s) 人类核心边缘价值（The Human Edge）与伦理思考 当 AI 融入日常工作后，对话的重点开始向“人类独特优势（Human Edge）”转移。员工们不再担心被算法取代，而是开始主动思考如何对 AI 的输出结果提出质疑，甚至有非科研背景的普通员工提出了深刻的伦理问题：“在 AI 自动化流程中，究竟谁才是最大的受益者？”这种转变标志着领导力与伦理思维在基层员工心中的萌发。 调查显示，89% 的员工明确表示，在 AI 深度嵌入的世界里，他们自身岗位的价值和重要性反而上升了。员工能够清晰识别出那些无法被算法替代的核心人类技能——包括批判性判断力、伦理决策能力以及复杂协同能力。 - **核心数据**：89% 的员工认为在 AI 时代其岗位重要性显著提升。 - **关键人类技能**：人类判断力（Judgment）、伦理决策（Ethical decision-making）、人际协作（Collaboration）。 - **转变本质**：从被动接受技术输出，转变为主动进行伦理审查与算法质疑。 --- ### [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) 五阶段人本 AI 应用框架与核心经验总结 基于三年的实践，Biscoe 总结并提出了一套可复制的“五阶段人本 AI 应用框架（Five-Stage Framework）”，为各类组织的管理者提供了明确的落地指南。她特别强调，阶段二（需求研判）与阶段五（聚焦人类优势）是绝大多数领导者最容易忽略、却最为关键的环节。 ```mermaid flowchart TD accTitle: Five Stage Human Centered AI Adoption Framework subgraph Phase1["基础奠定"] S1["阶段一：领导层承诺与学习示范"] --> S2["阶段二：需求调查与事前研判"] end subgraph Phase2["广泛推广"] S2 --> S3["阶段三：全院级认知与统一词汇"] --> S4["阶段四：特定岗位深度定制培训"] end subgraph Phase3["价值升华"] S4 --> S5["阶段五：确立人类边缘优势"] end ``` 针对实践中的经验教训，Biscoe 总结了五大核心启示与六项落地建议： #### 五大核心启示 1. **成人学习挑战**：AI 应用本质上是成人学习与行为改变的问题，而非纯粹的 IT/技术项目。一旦将其归类为 IT 项目，便会失去大众的参与度。 2. **领导力加速作用**：领导层的深度参与和主动学习能够极大加速全员的应用进程。 3. **正视恐惧心理**：必须承认并正面处理员工的焦虑，恐惧是起点而非阻碍。 4. **实践应用高于一切**：唯有结合日常工作的实操才能真正打动员工。 5. **人类技能是战略资产**：沟通、伦理与判断力等“软技能”并非可有可无，而是组织在 AI 时代最核心的竞争壁垒。 #### 管理者落地建议 - **先调研后设计**：在开发任何培训课程前，务必先开展员工调查。 - **构建通用素养**：从最基础的 AI Literacy（AI 素养）切入，打造团队的共同语言。 - **创造共享体验**：通过大型全员活动建立集体认知与讨论氛围。 - **岗位专项深耕**：针对会计、HR、研究员等不同岗位提供定制化培训。 - **健全伦理框架**：在技术应用的同时建立清晰的伦理与风控标准。 - **双线并行发展**：在提升技术硬实力的同时，同步开发人类核心软技能。 --- ### [16:00](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=960s) 研究局限性与未来跨行业合作展望 在演讲的尾声，Biscoe 保持了严谨的学者态度，客观分析了本次研究的局限性，并向在场的学术界、企业界及非营利组织发出了合作邀请。她指出，虽然俄克拉荷马大学的案例取得了显著成效，但目前研究仅基于单一机构，其结论尚未具备普遍的推广性（Not yet generalizable）。 此外，调查问卷的参与属于自愿性质，可能存在样本选择偏差（即填写者多为积极性较高的员工）；同时，过去三年间使用的调查工具和指标随时间有所演变。目前团队正在构建更加标准化、纵向（Longitudinal）的跟踪研究体系，希望能回答更多深层次的问题。 - **待解答的核心科研课题**： 1. AI 的广泛应用在长期内究竟如何影响员工的生产力与心理健康（Well-being）？ 2. 哪些具体的人类技能与成功的 AI 深度融合存在最强的正相关性？ 3. 该框架在企业、政府及非营利组织等不同环境下的适配与复刻效果如何？ - **合作邀请**：开放学术论文全文及提炼的 Key Takeaways，寻求多中心对比研究（Multi-site study）及企业跨行业合作伙伴。 ## Notable quotes > [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) AI 落地停滞不前的首要原因从来不是技术本身，而是组织忽视了人类心理与行为转变的规律。 — Belinda Biscoe > [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) 最强大的成果不是大家更多地使用 AI，而是他们更加清楚是什么让自己不可替代。 — Belinda Biscoe > [00:04](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=4s) 最终胜出的组织不是那些拥有最好工具的组织，而是那些率先解决‘人’这一维度的组织。 — Belinda Biscoe > [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) 这项技术违背了上帝与人性（This is against God and humanity）。 — 问卷受访员工 > [02:16](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=136s) 恐惧并非失败，而是变革的起点。 — Belinda Biscoe > [10:21](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=621s) 在 AI 自动化流程中，究竟谁才是最大的受益者？ — 学院员工 > [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) AI 应用本质上是成人学习与行为改变的问题，而非纯粹的 IT/技术项目。一旦将其归类为 IT 项目，便会失去大众的参与度。 — Belinda Biscoe > [11:49](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=709s) 沟通、伦理与判断力等‘软技能’并非可有可无，而是组织在 AI 时代最核心的竞争壁垒。 — Belinda Biscoe > [16:00](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk&t=960s) 虽然俄克拉荷马大学的案例取得了显著成效，但目前研究仅基于单一机构，其结论尚未具备普遍的推广性。 — Belinda Biscoe ## Mentioned Belinda Biscoe, 俄克拉荷马大学 (The University of Oklahoma), 外展与继续教育学院 (College of Continuing Education / Outreach), Ai4, Microsoft Copilot, 外展州况大会 (State of Outreach), 五阶段人本 AI 应用框架 (Five-Stage Human Centered AI Adoption Framework) ## Video description Belinda Biscoe, Senior Associate Vice President for Outreach/College of Continuing Education, The University of Oklahoma I've watched an organization of 1,400 people go from fear to fluency with AI, and I can show you exactly how we did it. Whether you lead a Fortune 500 team, a startup, or a university, the #1 reason AI adoption stalls isn't the technology; it's the people, and most leaders are skipping the steps that actually change behavior. In this session, I'll share three years of real data, a five-stage framework you can take back to your organization Monday morning, and the counterintuitive truth we discovered: the most powerful outcome of our AI initiative wasn't that people used AI more; it's that they became clearer about what makes them irreplaceable. Come ready to rethink your AI strategy, because the organizations that will win aren't the ones with the best tools; they're the ones who figured out the human side first. [Source](https://www.youtube.com/watch?v=J4Q4sOQ7Uqk)

Source quotes

[0:04] Belinda Biscoe: AI 落地停滞不前的首要原因从来不是技术本身，而是组织忽视了人类心理与行为转变的规律。 [0:04] Belinda Biscoe: 最强大的成果不是大家更多地使用 AI，而是他们更加清楚是什么让自己不可替代。 [0:04] Belinda Biscoe: 最终胜出的组织不是那些拥有最好工具的组织，而是那些率先解决‘人’这一维度的组织。 [2:16] 问卷受访员工: 这项技术违背了上帝与人性（This is against God and humanity）。 [2:16] Belinda Biscoe: 恐惧并非失败，而是变革的起点。 [10:21] 学院员工: 在 AI 自动化流程中，究竟谁才是最大的受益者？ [11:49] Belinda Biscoe: AI 应用本质上是成人学习与行为改变的问题，而非纯粹的 IT/技术项目。一旦将其归类为 IT 项目，便会失去大众的参与度。 [11:49] Belinda Biscoe: 沟通、伦理与判断力等‘软技能’并非可有可无，而是组织在 AI 时代最核心的竞争壁垒。 [16:00] Belinda Biscoe: 虽然俄克拉荷马大学的案例取得了显著成效，但目前研究仅基于单一机构，其结论尚未具备普遍的推广性。

Source quotes

Belinda Biscoe: I've watched an organization of 1,400 people go from fear to fluency with AI, and I can show you exactly how we did it. Belinda Biscoe: Whether you lead a Fortune 500 team, a startup, or a university, the #1 reason AI adoption stalls isn't the technology; it's the people, and most leaders are skipping the steps that actually change behavior. Belinda Biscoe: the most powerful outcome of our AI initiative wasn't that people used AI more; it's that they became clearer about what makes them irreplaceable. [0:04] Belinda Biscoe: Leading Human-Centered AI Adoption: From Awareness to the Human Edge. [2:16] Survey Respondent: This is against God and humanity. [4:53] Belinda Biscoe: Oh, this is so exciting. [7:48] Staff Member: Who benefits more from an AI-automated process? [15:36] Belinda Biscoe: The organizations that will win with AI in this era will not be the ones with the best tools; they will be the ones who figured out the human side first.

Original transcript

[0:04] I'm Belinda Biscoe, the senior associate vice  president for university outreach in the college [0:11] of continuing education at the University of  Oklahoma. We're one of the nation's largest [0:17] continuing education organizations, and it's there  that I lead about 1,400 employees. But today, what [0:27] I want to do is to talk to you about our journey  into AI, representing and presenting it as a case [0:36] study entitled leading human-centered AI adoption  from awareness to human to the human age. And what [0:47] I'm really hoping is that this will resonate with  some of you. I know that we have academics here, [0:54] we have corporations, nonprofits, industry people,  and I think a bunch of what we've learned may be [1:01] applicable across many sectors. Let me start by  saying that in 2024, 23-24, half of my team at the [1:13] university had never experienced or touched AI.  Some of them were terrified, some were furious, [1:23] and a few were even convinced that their careers  would end. I'm happy to say that now in 2026, [1:32] 95% of them can tell you exactly what makes them  irreplaceable in an AI-embedded world. And I [1:43] didn't do this by investing first in technology,  but by first investing in people. So, if some of [1:52] you are experiencing what I did initially  with people being fearful and frightened, [1:58] and your AI adoption stalling. I want to talk  to you today about how we overcame that and give [2:07] you a five-stage framework that hopefully you can  take back to your organizations tomorrow. So, how [2:16] many of you have individuals in your organization?  Raise your hand if you're leading an organization [2:23] and you have people who are afraid of AI still.  If you're Yeah, there's still a lot of people [2:30] out there. How many of you have purchased an AI  tool and your team still isn't ready to adopt it? [2:39] Some hands. Well, like Robbie, our  previous speaker, I'm a baby boomer, [2:45] but I tend to be an early adopter. So, 5 years  ago, when I began experimenting and exploring AI, [2:54] I knew that this would be a game-changer. And  it was at that moment that I decided, "Belinda, [3:02] you have got to begin to bring AI literacy to the  entire college." There was so much fear. The first [3:12] thing I did was give a college-wide survey,  and believe it or not, there was one person, [3:19] and I say this is a direct quote, who said on  the survey, "This is against God and humanity." [3:28] So, knowing that AI exist is one thing.  People went from fear and paralysis. [3:37] There was lots of resistance, but again, I  am here to say that we went beyond that. So, [3:48] if you look at the data there, you will see that  we are one of the largest continuing education [3:56] units. Again, we have about 1,400 people.  About 800 of those are full-time people, 600 [4:04] are part-time and occasional workers. And we did  this over about 3 years. We have 17 departments. [4:12] But before I get into that, I want to say a  little bit about the methodology. One of the [4:18] things that we did was to use what we call a  mixed methods study over this 3-year period. [4:26] We distributed surveys in year one to everyone, to  all of our employees, and we provided professional [4:35] development opportunities for everyone. We did  that across the 3 years. And so, what I want to [4:45] to say about that is that it is really  important to know where you're starting. [4:53] So, here is our journey at kind of  a glance. We went from awareness, [5:01] we had a foundation there, we then went  to application, and then we went to [5:10] actually the human edge, where we started to focus  on those things that we think are not replaceable. [5:22] And so, if you look at the slide that's in front  of you, we acknowledge that the the fear was real. [5:32] AI people said as a huge threat to my job. There  were people who absolutely said they would quit [5:41] working at the college if we went forward.  These were not fictitious people. These were [5:48] real people. You can see that at the end of that,  there were in fact in that first year at least 28 [5:58] people, 28% who felt that there were no meaningful  benefits to AI. But, what I want to say is this is [6:07] what we did not do. We did not dismiss the  fear. We didn't not send an email saying, [6:16] "Oh, this is so exciting." We embraced all  the problems. We did not rush past the fear, [6:23] and we created psychological safety. I want  to say to you, fear is not failure. It is a [6:30] starting point. We acknowledge that AI, like most  new innovations and inventions, brings with it [6:39] both strengths and challenges. If you look at what  is on the slide now and the approach that we use, [6:50] every year we have a large event. We call it our  state of outreach, and it was during this time [6:57] that we used that as an opportunity to develop  and to present huge and a number of professional [7:06] development opportunities. This strategy  allowed us to emphasize organizational learning, [7:14] workforce readiness, ethical reflection, and  practical application. I want you to really [7:24] make note of that because the professional  development was intense, it was ongoing, [7:31] and it was not a one-time endeavor. We then  also provided in one of those years an in-depth [7:41] Microsoft Copilot deep dive  training on a voluntary basis. [7:48] The most important thing is that this training  went on every year with more and more in-depth. [7:56] The important thing here also is that we ran  surveys before we designed the programming, [8:04] before not afterwards. I as a leader of that  organization also embraced curiosity and interest. [8:15] It was really important that people understand  that I too was a learner, a lifelong learner. [8:25] If you look at what is there now, let me show  you and tell you what one year of investment in [8:34] our people actually did. In one year without a  new AI platform, no new mandates, no coercion, [8:43] we went from 50% of our people having zero  experience to about 74% using AI regularly [8:53] or occasionally. I want you to know that the  concerns did not disappear, but they matured. [9:02] We had a shift from existential  fear to operational sophistication [9:09] and I consider that organizational  growth. We had people evolving concerns of [9:18] I felt that their concerns were no longer a  sign of resistance, but a sign of progress. [9:27] So, I want to be clear about what we're talking  about here. These were administrators, educators, [9:36] human resources professionals, customer service  representative representatives, grant writers, [9:43] people just like the people on many of your teams.  You can see some of the things that people began [9:51] to do and the things they began to use. You  can also see that the issue was for me about [10:01] the sequencing of things. I think often in our  organizations the barrier is the launch sequence. [10:11] We used a structured intentional people-first  approach and now that is what I'd like to say [10:21] a little bit more about. You can see that  this year with our focus on the human edge [10:29] that we literally had people who were asking  questions about whether or not there were log [10:42] algorithms in things and wondering how they  can question the outcomes from AI outputs. So [10:52] people move from fear to really realizing that the  human aspect of this really, really was important. [11:02] They could identify key human skills, judgment,  ethical decision-making, collaboration. [11:11] 89% could even tell you that their roles were  even more important in an AI-embedded world. [11:23] The conversation evolved over time and  I love the last one up there and that is [11:31] some people were asking, "Who benefits more  from an AI-automated process?" You can see that [11:40] this for me reflected more ethical leadership  thinking, and this was from a non-researcher. [11:49] Here is the five-stage framework that  I want to say a little bit more about. [11:56] Stage one, and I think this is really important  for organizations. Leadership must make a [12:03] commitment, and obviously demonstrate interest  in the learning. This isn't optional. Stage two [12:13] is really critical. As a research psychologist,  I realized that we needed information. [12:20] What did our employees know? What did  they fear? And what did they need? [12:26] The answers will surprise you. We constantly  surveyed before every professional learning [12:35] and after professional learning, and we geared our  training and learning based on that. Stage three, [12:44] organizational-wide awareness. We wanted to build  a common vocabulary. We wanted to create shared [12:53] experiences. We wanted to bring everyone together  around the same conversation. And then, this is [13:01] also equally important. As we took a deeper dive,  we began to do role-specific training. What does [13:11] an accountant need? What do our human resources  people need? What do our researchers need? [13:18] Because as an adult learner myself, and as someone  who studies adult learning, we knew that people [13:26] would become more connected, more engaged if the  learning was specific to what they do every day. [13:34] So, I point out to you that stages two and five  in my framework are critically important, and I [13:43] think sometimes that those are the stages that  we as leaders overlook. There are some lessons [13:52] that I think are worth taking for this. One, I  do believe that AI adoption is an adult learning [14:02] challenge and not a technology challenge.  The moment we say this is an IT project, [14:09] I think we've lost our audience. Lesson two,  leadership engagement accelerates adoption. [14:18] Three, fear must be dealt with and acknowledged.  Uh and practical application is everything. [14:27] Finally, I think it's really important to  note that human skills are not soft skills, [14:34] they are strategic assets. And in an AI-enabled  world, I believe that they will become your [14:44] organization's greatest competitive edge. So, what  are the recommendations I want to leave with you? [14:54] Survey before you design, start with basic AI  literacy, create shared experiences, invest in [15:05] role-specific training, build ethical frameworks,  and develop human skills alongside technical ones. [15:16] I truly believe that the organizations that will  win with AI in this era will be the ones will not [15:25] be the ones with the best tools, but they will be  the ones that figured out the human side first. [15:36] And the organizations that win in the AI era may  not have the best tools, but they will have the [15:44] best people using them. I really believe this is  whether we're talking higher education or other [15:53] sectors. This is what we built at the University  of Oklahoma Outreach College of Continuing [16:00] Education. One of the things I want to say  as I studied and reflected on what we did [16:10] is that there's still some things we don't know,  and I really want to open the conversation up to [16:17] those of you who are here. We didn't replicate  and we haven't replicated this in other settings, [16:24] nonprofits, corporate settings, government  settings. We would love to have conversations with [16:31] some of you about partnering with your sectors. We  also don't know, for example, how does AI adoption [16:41] affect employee productivity and well-being  over the long term. These are research questions [16:48] that we would like to investigate and which  specific human skills correlate most strongly [16:55] with successful AI integration. These are the  good questions that keep me up at night. There [17:03] also some honest limitations to what we did. This  was one institutions. While we think our findings [17:12] are compelling, they are not yet generalizable.  Survey participation was voluntary, which means [17:20] we probably had motivated employees taking it. Our  survey instruments evolved over time and we used [17:28] different instruments each year. And so, we are  building toward a more standardized longitudinal [17:37] design. And so, many of you in this room represent  organizations where the research could go next. [17:45] If you're interested in replicating this framework  in a comparative multi-site study or longitudinal [17:54] tracking of AI adoption in your workforce, I  would love to talk to you. My information is [18:04] on this slide. The full paper that I presented  for submission for the academic day proceedings [18:12] is available. And I have some key takeaways there  for you that I think I have already mentioned. [18:23] And so, thank you for your time. I have enjoyed  being here with you. And I hope there were [18:30] a few things here that might have been of  interest to you and your organization. Thank you.