Voltava Deploys Isolated Laptops to Drive Manufacturing AI Shift

Voltava部署隔离笔记本推动制造业AI转型 工业供应商Voltava利用隔离沙箱环境与平民开发者设备，并结合员工自然减员，在保护企业网络的同时加速AI应用普及。 在全球拥有21个中心的大型一级制造商Voltava的高管们探讨了将AI与人力战略相结合并融入工业运营的方法。该领导团队在快速创新与技术治理之间取得平衡，利用沙箱环境、领域大使和平民开发者笔记本电脑绕过传统数据瓶颈。Voltava并未谋求裁员，而是借助效率提升与自然减员来增强人类认知能力，同时倡导跨行业合作以应对共同面临的运营与技术基础设施难题。 全球一级制造巨头Voltava在全球运营21个制造中心，拥有超5,000名员工，每年为汽车和国防等行业生产超过3,000万套复杂零部件总成。 (0:04) CEO Bruce Smith与CTO Tom Swartz构建了“油门与刹车”式的领导协作机制，在建立包容质疑的心理安全文化的同时，平衡AI前沿创新与技术治理安全。 (2:08) 为防范核心网络风险，Voltava利用独立实体Kinetic作为多智能体与CRM自动化的试验沙盒，并在防火墙外配备了20台公民开发者笔记本及15位跨部门大使。 (5:46) Voltava避开旷日持久的传统数据清洗困局，通过在7条创新赛道中结合合成数据与弹性模型并行推进，并依靠自然减员而非裁员实现以人为本的产能跃升。 (9:57)

Executives at Tier-1 manufacturer Voltava argue that successful industrial AI adoption depends on pairing entrepreneurial ambition with rigorous technical veto power, workflow restructuring, and safe experimentation. Panelists Bruce Smith, Tom Swartz, and Greggory Garrett emphasize that sustainable transformation requires cultural psychological safety, pragmatic data integration, and cross-industry cooperation on shared infrastructure constraints. Executive Vision Counterbalanced by Technical Governance (2:08) Operating 21 global manufacturing centers and producing over 30 million assemblies annually, Voltava manages technology adoption through a deliberate division of executive roles. Owner-operator Bruce Smith functions as the accelerator driving adoption, while technical leader Tom Swartz holds explicit veto authority to discard initiatives lacking viable architectural roadmaps. Smith notes that unchecked enthusiasm risks chasing technological fads, drawing historical comparisons to earlier operational shifts such as late-1980s industrial robotics, lean manufacturing, and the early corporate rollout of laptops. This operational balance is underpinned by psychological safety that encourages direct internal challenge. Smith and moderator Greggory Garrett observe that managing agentic systems shifts executive value toward high-order problem formulation rather than baseline execution, warning that organizations must sustain candid debate around operational limits to avoid flawed deployments. Sandboxed Experimentation and Pragmatic Data Maturation (5:46) To innovate without exposing core enterprise networks to risk, Voltava isolates rapid prototyping and multi-agent CRM experimentation within Kinetic, a non-automotive sister entity. Inside Voltava proper, the company deployed 20 standalone 'citizen laptops' outside the corporate firewall to let staff test agentic tools, supported by approximately 15 departmental ambassadors who bridge operational floor realities with IT engineering. Swartz explicitly rejects standard enterprise guidance that demands delaying AI deployment until all legacy data pipelines are thoroughly scrubbed. Instead, Voltava advances data capabilities across seven concurrent operational swim lanes, utilizing synthetic data and adaptable models to progress in parallel rather than deferring projects for multi-year data cleanup efforts. Workflow Redesign and Workforce Cultural Alignment (9:57) Voltava structures its AI deployment around organizational redesign rather than head-count reduction, combining efficiency gains with natural attrition. Smith outlines a human-centric approach that frames technology as an employee tool, actively countering workplace anxiety by drawing internal skeptics directly into strategic discussions rather than punishing dissent. Garrett introduces the concept of the 'minimum viable human organization,' where workforce roles transition away from mechanical execution toward framing and solving complex challenges. The panelists state that automating preexisting steps in fragmented supply chains only cements underlying friction. Long-term productivity gains require leadership to rethink entire operational handoffs and incorporate frontline pushback into system design from the outset. Industry Collaboration on Infrastructure and Governance (14:52) Swartz urges AI vendors and manufacturing partners to look beyond conceptual pitch decks and deliver production-ready applications for shop-floor, engineering, and finance environments. Because global manufacturing remains heavily fragmented by decades of verticalization and international component handoffs, participants argue that firms must coordinate on systemic challenges rather than treating every implementation as proprietary. To address common technical obstacles—such as autonomous agent safety, secure data wiping, and physical constraints like regional power availability and data center capacity—Swartz participates in customer-led advisory councils alongside direct market competitors. Voltava views collective governance and shared industry baselines as mandatory requirements for navigating the sector's operational transition. The panel concludes that AI's impact on manufacturing hinges far more on organizational redesign, psychological safety, and clear problem articulation than on software capabilities alone. Unresolved questions remain regarding how the broader industrial ecosystem will overcome shared physical constraints, particularly power grids and data center capacity, while establishing common governance standards for autonomous agents.

全球一级制造供应商 Voltava 高管指出，工业人工智能转型的核心不在于跟风引入技术工具，而在于建立领导愿景与工程治理相互制衡的决策机制，并通过流程重构、沙盒隔离和包容异见的组织文化推进人机协同。 领导权责制衡与技术防过热机制 (2:08) 作为在全球拥有21处制造中心、逾5000名员工且年产超3000万套复杂组件的总成供应商，Voltava 的高层采取了一种结构化的决策制衡模式。企业所有者兼运营者 Bruce Smith 自述其对新兴技术的探索热情极易引发内部震荡，因此他赋予技术负责人 Tom Swartz 针对各项 AI 创议的全权审核与一票否决权，以防企业追逐缺乏可行落地路径的虚假热点。双方将该协作机制比喻为车辆的油门与刹车：Smith 负责推动技术的敏捷采纳，Swartz 则负责把控网络安全、系统架构与合规治理的底线。 Smith 认为，当下的 AI 浪潮与其职业生涯中见证过的1980年代末期工业机器人引入、日本精益生产模式普及以及企业便携电脑应用属于同等量级的历史拐点。为了应对技术变革带来的不确定性，管理层在内部建立起允许公开推敲与反对的心理安全文化，明确技术推进必须建立在严谨的技术验证之上，而非单纯依赖概念宣传或外部狂热。 隔离沙盒与去中心化落地架构 (5:46) 为了在保护核心制造网络安全的同时鼓励创新，Tom Swartz 实施了一套多层级的技术落地架构。该体系首先依托非汽车业务的关联实体 Kinetic 作为前沿试验平台，在完全隔离的环境下进行多智能体工作流、CRM 自动化实验以及新型装配工艺的原型验证。在 Voltava 主体架构内，公司选拔了约15名来自各业务部门的领域大使，直接建立起日常生产运维与 IT 核心开发团队之间的沟通桥梁。 针对基层员工的探索需求，Voltava 配发了20台处于企业防火墙之外的专用笔记本电脑，供业务人员开展低代码与智能体开发实验。Swartz 明确反对必须耗费数年彻底完成遗留数据清洗才能部署 AI 的传统论调；相反，Voltava 设立了7条平行的业务创新赛道，借助合成数据与灵活性模型推进方案落地，在技术实践与验证过程中同步提升数据资产的成熟度。 人机密度优化与反对意见包容 (9:57) 在人机协同与劳动力演进层面，Bruce Smith 主张以人为中心的技术伦理，明确提出管理层应当善用技术服务于员工，而非将员工视为技术部署的耗材。Voltava 不采取因技术替代而导致的激进裁员方案，而是结合员工自然减员与 AI 带来的生产率提升来实现业务现代化的平稳过渡。CGS Advisors 首席执行官 Greggory Garrett 对此提出了“最小可行人类组织”的概念，认为在智能体时代，企业领导者的核心职能正在从基础执行转向高层级的问题界定与架构设计。 为缓解员工面对大语言模型等新兴工具时的岗位焦虑，管理层提倡保持彻底透明的沟通，并主动接纳内部质疑。在面对员工对大语言模型可靠性的尖锐质疑时，高管团队选择展开公开对话而非压制异见。Garrett 总结指出，推进组织转型必须尽早将持怀疑态度的员工纳入决策过程，将潜在的阻力转化为技术落地时的校验与纠偏机制。 流程实质重构与跨行业生态协同 (14:52) Greggory Garrett 指出，汽车与国防制造领域在数十年纵向分工下高度碎片化，跨国采购与频繁交接造成了严重的流程摩擦。因此，AI 的真正价值在于促使管理者审视全局、消除不必要的冗余工序，而非仅仅将原有低效步骤予以自动化。Swartz 呼吁技术供应商与行业伙伴放弃泛化的宣传方案，转向能直接集成于供应链、财务和车间生产环境的高可用场景。 随着智能体系统的深入部署，制造企业普遍面临自主智能体行为约束、数据擦除标准以及电力和算力基础设施短缺等共有挑战。为此，Voltava 参与了由客户主导并联合直接竞争对手的行业顾问委员会，共同制定跨行业的数据治理规范与安全框架。Swartz 与 Smith 强调，制造业 AI 转型并非单一企业的孤立工程，而是一场依赖跨机构信任与深度生态协同的系统化重塑。 Voltava 的实践表明，传统制造企业的智能化转型高度依赖于决策制衡、流程剔除重塑以及跨供应链的标准协同。然而，正如讨论所提及，自主智能体在实体工业场景中的行为约束、遗留数据管道的动态治理，以及区域电力与算力基础设施的物理瓶颈，依然是全行业必须共同面对并待解的现实制约。

## Background 该内容由 Ai4 于 2026 年 8 月 23 日发布在 YouTube 平台上。视频记录了 CGS Advisors 首席执行官 Greggory Garrett 与全球 Tier 1 制造巨头 Voltava 的董事长兼 CEO Bruce Smith 及 CTO Tom Swartz 展开的炉边对谈，深入探讨制造业如何打破传统生产力范式、构建现代化的“智能战略（Intelligence Strategy）”。 ## TL;DR 对话围绕拥有 21 个全球制造基地的制造业巨头 Voltava 的 AI 落地实践展开。两位高管分享了独特的“油门与刹车”协作模式，即 CEO 赋予 CTO 51% 的一票否决权，并通过独立的 Kinetic 沙盒、15 位业务领域 AI 大使以及物理隔离的“公民笔记本”安全推进基层创新。Tom 破除了必须先花巨资清洗历史数据的传统迷思，倡导并行多泳道与合成数据应用；Bruce 则强调“爱护员工、善用技术”的人机乘数效应，依托自然减员与心理安全感消除员工恐慌。整体内容指明企业必须完成从孤立的“AI 工具战略”向重构组织工作流与决策机制的“全局智能战略”跨越。 ## Atmosphere & style 整场炉边对谈务实接地气且充满思辨张力，既有老牌制造掌舵人经历数十年技术周期后的通透与沉稳，又有技术高管深耕车间一线的敏锐与严谨。CEO 与 CTO 之间的言语互动坦诚直率，完美呼应了他们所推崇的“痛快争吵后再紧紧拥抱”的求真文化。讨论摒弃了空洞的技术炒作，以极具组织安全感的真实案例直击制造业转型的深层痛点，呈现出一种充满信任感、务实求真的实战交流氛围。 ## Key takeaways - [00:04](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=4s) Voltava 作为拥有 21 个全球基地、年产超 3000 万套精密组件的 Tier 1 制造巨头，正积极探索从传统制造业向现代化智能战略的深度转型。 - [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) CEO 与 CTO 之间建立了“油门与刹车”的高效协作机制，CEO 主动赋予 CTO 51% 决定权与一票否决权，确保前瞻性构想具备落地可行性与安全性。 - [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 在安全的组织环境中，建设性冲突能够高频发生；团队需要学会“有分歧但不失和气”，秉持做正确的事而非争当正确的原则。 - [06:03](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=363s) 通过设立独立非汽车制造企业 Kinetic 作为创新沙盒，Voltava 能够在不影响核心业务的前提下实验自动化 Agent 与多智能体协同机制。 - [06:03](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=363s) 通过业务部门选拔的 15 位“AI 大使”与配发物理隔离设备的“公民笔记本计划”，打通业务痛点并赋能基层员工进行低代码/零代码实验。 - [08:17](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=497s) 破除必须等数据完美无瑕才开始做 AI 的误区；Voltava 采用 7 条并行成熟度泳道推进场景落地，避免在别人下国际象棋时原地玩跳棋。 - [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 人类高阶认知与机器执行智能之间是“乘数效应”而非简单加法，关键在于找准人机分工密度以释放指数级商业价值。 - [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 奉行“爱护员工并善用技术，绝不崇拜技术而利用员工”的哲学，借助自然减员和技能重塑消除裁员恐慌，与员工携手并进而非强加变革。 - [13:36](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=816s) “最小可行人类组织”（MVHO）模式下，领导者的核心竞争力正在转向对复杂业务场景的“问题表述与框架界定”。 - [14:52](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=892s) 制造业需要打破行业孤岛，与能够派遣工程师一线落地的伙伴合作，甚至与竞争对手共同制定 AI Agent 跨系统访问和数据治理准则。 - [17:00](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1020s) 企业转型必须从局部的“AI 战略”升维到全局的“智能战略”，同步涵盖高管思维转变、组织架构重塑和业务工作流重构。 - [18:51](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1131s) 建立高度的心理安全感与求真文化，倡导“痛快吵上一架再拥抱全速前进”（slug it out and hug it out），依靠真实坦诚应对转型挑战。 ## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=4s) 制造业巨头 Voltava 与 AI 转型的破局者 主持人 Greggory Garrett（CGS Advisors 首席执行官兼董事总经理）作开场致辞。CGS 在过去 15 年间一直致力于协助企业领导者应对技术颠覆与组织转型，在物联网（IoT）和工业 4.0 领域积累了深厚经验。针对当下的 AI 浪潮，CGS 调研了数百位高管（C-Level Leaders），并将相关洞察融入到本场对话中。随后，Greggory 介绍了本场的对话主角——全球制造巨头 Voltava。作为汽车与国防等行业顶级的 Tier 1 级零部件组装制造企业，Voltava 拥有 21 个全球制造中心、超 5,000 名员工，每年交付多达 3,000 万套精密复杂的组件产品。 舞台上落座的两位嘉宾分别代表了战略决策与技术落地的核心力量：一位是 Voltava 的董事长、CEO 兼企业所有者 Bruce Smith，作为极少数具备“所有者兼运营者（Owner-Operator）”视角的制造业掌舵人，他带来了对 AI 价值创造的现实思考；另一位则是 Voltava 的首席技术官 Tom Swartz，他承担着将 Bruce 的宏大愿景转化为落地现实，并确保技术具备可扩展性与高安全性的重任。 - **Voltava 企业规模**：全球 21 个制造基地，员工超 5,000 人，年产各类高复杂度总成组件超 3,000 万套。 - **产业定位**：汽车制造、国防装备转型等高精密领域的头部一级供应商（Tier 1 Assembler）。 --- ### [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 领袖与技术官的双人舞：油门、刹车与 51% 否决权 当被问及制造业 CEO 应当如何看待 AI 并不被虚火迷惑时，Bruce 直言不讳地指出，领导者的兴奋对组织来说既是“恩赐也是诅咒”——老板一旦表现得过度亢奋，整家公司都会引发剧烈的连锁震荡。因此，他奉行的核心理念是“合作是成功者的秘诀，而非弱者的拐杖（Partnerships are the secret of the successful, not the crutch of the week）”。在与 CTO Tom 的协作中，Bruce 主动赋予 Tom **51% 的决定权与一票否决权**。Bruce 自认擅长提出极具前瞻性的宏大构想，但他需要 Tom 负责绘制可行的技术路线图，并在繁杂的想法中“去芜存菁，辨别沙砾与真金（separate the glitter from the gold）”。 ```mermaid flowchart LR accTitle: 战略与落地的协作飞轮 subgraph S1["CEO 愿景驱动"] A["战略远见与业务构想"] --> B["加速推动（踩油门）"] end subgraph S2["CTO 落地治理"] C["技术路线与安全评估"] --> D["51% 否决权（踩刹车）"] end subgraph S3["最终决策闭环"] E["高频安全建设性冲突"] --> F["做正确的事而非争当正确"] end B --> E D --> E ``` Bruce 强调，健康的伙伴关系依赖于极高的组织安全感。他坚信“当环境足够安全时，冲突就会高频发生（When the environment is safe, conflict is frequent）”，团队应当学会“有分歧但不失和气（disagreeing without being disagreeable）”。他将两人的配合比作一辆性能完备的汽车：“世上没有任何一辆汽车只有油门或只有刹车；我总是想开得更快（踩油门），Tom 总是想更稳健审慎（踩刹车），懂得在何种场景下如何配合施加力度，才是既安全又迅速抵达目的地的艺术。”谈及为何早在数年前就积极拥抱 AI，Bruce 回顾了自己职业生涯中经历的数次技术拐点——从 80 年代末的工业机器人普及，到日本精益制造席卷全球，再到企业首次引入笔记本电脑。在他看来，AI 是一场永久重塑工作方式的根本性变革，正如他告诫团队的名言：“我们可以忽视人工智能，但我保证，它绝不会忽视我们。” --- ### [06:03](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=363s) 敏捷沙盒与组织赋能：Kinetic 试验田、AI 大使与公民笔记本 CTO Tom Swartz 详细分享了 Voltava 在承接 CEO 愿景时的组织与技术架构设计。为了在不影响核心汽车制造业务稳定性的前提下探索创新，Bruce 成立了一家名为 **Kinetic** 的独立非汽车制造企业。Kinetic 顺理成章地成为了前沿技术的“创新沙盒”，团队在此实验新型制造流程、开发自动化 Agent（智能体）、构建多智能体协同机制（Agent-to-Agent），并将其成功部署在 CRM 等关键业务场景中。 ```mermaid flowchart TD accTitle: Voltava AI 创新与孵化三层架构 subgraph Box1["外部试验田：Kinetic 沙盒"] K1["非汽车独立业务单元"] --> K2["探索创新产品与装配逻辑"] K2 --> K3["孵化多 Agent 协同与 CRM Agent"] end subgraph Box2["中枢连接层：15 位 AI 大使"] A1["业务部门领域专家 SME"] --> A2["梳理业务流程与痛点场景"] A2 --> A3["联合 IT 与工程师推向生产线"] end subgraph Box3["基层创新层：20+ 公民笔记本"] C1["隔离网络物理机"] --> C2["低代码 / 零代码开发与实验"] C2 --> C3["合规培训与优秀用例回流"] end ``` 在 Voltava 主体内部，Tom 推动了两项极具开创性的赋能机制： 1. **AI 大使机制（AI Ambassadors）**：约一年半前，公司在各业务部门选拔了约 15 位业务骨干作为领域专家（SMEs）。他们深谙本部门的底层流程与业务链路，负责挖掘高价值用例并对接 IT 研发团队，共同将 AI 原型推进至生产环境。 2. **公民笔记本计划（Citizens' Laptops）**：为满足约 20 位对 AI、低代码/零代码开发及自主搭建 Agent 充满热情的基层员工的需求，同时杜绝内部网络的安全与合规隐患，公司为他们专门配发了与企业内网物理隔离的独立笔记本设备。员工可在安全环境下接受培训与代码实验，一旦跑通高价值模型，再经过安全审计正式接入企业主网。 --- ### [08:17](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=497s) 破除“数据就绪困境”：拒绝在跳棋棋盘上打 AI 象棋 针对制造业普遍存在的“必须先花数百万美元和数年时间清洗历史数据才能做 AI”的传统论调（尤其是通过不断收购与整合拼凑起来的企业），Tom 坚决给出了否定答案。他尖锐地指出，如果一家企业必须等到数据完美无瑕才开始探索 AI，“那就像别人在下国际象棋，而你还在原地玩西洋跳棋”。数据固然是 AI 的基石，但数据本身是动态变化的，甚至未来会被各类 AI Agent 持续重塑；如果死守数据清洗，企业将永远错失发展窗口期。 Voltava 采取了并行推进的多泳道战略，将 AI 成熟度划分为 7 条并行泳道（包含战略路线图制定、治理框架、全员教育、组织共识等），在推进数据治理的同时同步进行场景落地。Greggory 也对此表示赞同，指出企业完全可以利用合成数据（Synthetic Data）打破现有的推理天花板（Reasoning Ceiling），在数据就绪之前通过模拟场景提前演练 AI 能力。 ### [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 乘数效应与“人与技术”的共生模式 讨论深入到企业如何真正让 AI 发挥实效的核心命题。当前技术的迭代速度极快，展厅里的各种新技术甚至已经超出了多数组织的消化能力，但真正的秘诀并不在于等待技术成熟，而在于让人类员工与技术并行协同。在通用人工智能（AGI）实现之前，企业必须采用混合模式。这种结合绝非简单的加法（人类智能 + 机器智能），而是一种“乘数效应”（Multiplicative）。关键在于找准“密度”（Density），明确界定人类与机器各自的角色分工，从而释放指数级的增长潜能。 ```mermaid flowchart TD accTitle: 人机协同乘数效应模型 subgraph Synergy["智能协作架构"] H["人类高阶认知<br/>(问题定义与业务构想)"] M["机器执行智能<br/>(AI Agent 与自动化流程)"] H -->|"乘数密度匹配 (×)"| S["指数级商业价值"] M -->|"乘数密度匹配 (×)"| S end subgraph Culture["组织文化保障"] P["爱护员工，善用技术<br/>(非利用员工、崇拜技术)"] O["自然减员与技能重塑<br/>(消除裁员恐慌)"] C["开放透明的对话机制<br/>(打破外界舆论信息差)"] P --- O --- C end Culture -.-> Synergy ``` 在谈及全球并购整合过程中的组织文化建设时，Bruce Smith 阐述了他的核心用人与技术哲学：“我热爱员工并善用技术，绝不崇拜技术而利用员工。”员工能敏锐察觉到管理层真正看重的是什么。Voltava 致力于为员工铺设平滑的过渡路径，形成重叠赋能的“维恩图”（Venn diagram），而不是单向的岗位替代。公司虽然常提醒员工“你不会被 AI 取代，但可能会被使用 AI 的对手取代”，但在内部落地各种 AI Agent 和生产力工具时，主要依托自然减员（Natural Attrition）而不填补空缺，从而避免员工产生“今天拥抱 AI，下周就会被解雇”的恐慌。 Bruce Smith 强调，AI 变革必须是“为了员工、与员工携手并进”（With and for our people），而不是“强加给员工”（To our people）。正如老话所说，“行动的声音太大，以至于人们听不清你说了什么”，领导层必须以实际行动兑现承诺。针对 AI 引发的未知恐惧，企业更要主动将 AI 搬上台面讨论，否则外界媒体的渲染会让员工把 AI 脑补成可怕的怪物。CTO Tom Swartz 分享了一个真实的案例：前几天有员工直言“大语言模型在公司毫无立足之地”，Tom 得知后立刻主动找该员工沟通交流，了解对方的顾虑所在，而没有采取任何惩罚或打击报复。这种允许公开表达异见且具有包容性的文化，让员工切身感受到公司是真正与大家站在一起的。 --- ### [13:36](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=816s) 最小可行人类组织与领导力新范式 在以蓝领工人甚至工会力量为主的传统制造业环境中，由 CEO 亲自带头公开推行 AI 具有极强的革新意义。当前的行业大会不仅需要技术讨论，更亟需设立专门针对首席人力资源官（CHRO）的议程，因为技术本身终会达标，而决定成败的真正瓶颈在于“人”如何融入这一变革。 关于人机分工，前沿研究提出了“最小可行人类组织”（Minimum Viable Human Organization, MVHO）的概念。这一理念绝非要淘汰人类员工，而是指组织应该专注于让人类去做只有人类才能胜任的高阶工作，并不断提升员工的技能维度。在 Agent 时代，领导者最核心的能力正在发生根本性转移——“问题的清晰表述与框架构建”（Problem articulation and framing）将成为首要竞争力，这标志着组织需求正从单纯的“执行工作”转变为“定义与拆解问题”。 - **人机交互理念**：技术不是替代劳动力，而是重构岗位密度，推动人类向高阶认知活动迁移。 - **领导力核心转型**：从指令型、事务型的执行管理，转向对复杂业务场景的“问题表述与框架界定”。 - **人力资源新挑战**：AI 转型需要组织架构、岗位职责与晋升体系的全面重组，CHRO 在制造业转型中扮演着前所未有的关键角色。 --- ### [14:52](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=892s) 拆除行业孤岛与跨界协作生态 转向具体落地层面时，Tom Swartz 梳理了 Voltava 在车间现场（Shop Floor）、后勤办公室（Back Office）、供应链以及财务等场景中对外部生态的核心诉求，并向现场的供应商、合作伙伴及竞争对手发出了开放协作的倡议。 ```mermaid flowchart LR accTitle: 制造业生态协同网络 subgraph Ecosystem["跨界协作生态"] V["供应商 / 开发者<br/>(落地车间与业务端实际用例)"] C["同业竞争对手<br/>(建立治理准则/数据安全共识)"] K["客户与合作伙伴<br/>(数据流打通/消除供应链断点)"] end Ecosystem ==> VT["Voltava 生产环境与智能制造平台"] ``` Tom Swartz 指出，虽然每天都有大量供应商在办公室门外排队，但 Voltava 真正需要的是那些拥有成熟落地案例、能够派遣工程师深入一线“脚踏实地”（Boots on the ground）解决实际制造问题的合作伙伴。与此同时，制造业必须打破长期以来的部门与企业壁垒（Silos）。即使在行业竞争对手之间，也需要展开坦诚乃至带有碰撞性的对话。目前 Tom 已代表公司加入了由核心客户主导的委员会，与诸多竞争对手围坐在一起，共同探讨在 AI Agent 跨系统访问、数据治理以及防止数据被恶意擦除等严峻挑战下的应对方案。面对数据中心与算力基建带来的行业重构，整个产业链必须共享用例与数据经验，携手应对转型期的共性难题。 汽车及高端制造供应链长期呈现高度碎片化和垂直分工的特点，各个环节之间存在大量繁琐的交接流转。引入 AI 的最大价值在于“拉远镜头”，以全局视角消除不必要的中间流程，而这一目标的达成高度依赖于全行业的协同与共建。 --- ### [17:00](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1020s) 从 AI 战略走向智能战略：高管思维的根本转变 各行各业推进 AI 转型的核心阻碍往往不在于技术本身的颠覆，而在于高管层思维模式（Executive Mindset）的根本转变。如果一个企业的 AI 转型项目缺少了组织架构重塑（Organizational Redesign）、业务工作流重构（Workflow Redesign），以及将持怀疑态度的阻力群体纳入共创机制的包容方案，那么该转型项目就存在致命的结构性缺失。 真正的转型不是孤立地制定一份“AI 战略”，而是要构建全局性的“智能战略”（Intelligence Strategy）。在这一体系中，技术的终极目标是赋能业务人员做出更精准、更具确定性的决策，实现全链路的价值释放。 | 转型维度 | 传统技术驱动模式 | 现代化“智能战略”模式 | |:--- |:--- |:--- | | **关注核心** | 单纯引入 AI 算法与工具 | 高管思维转型与人机协同密度设计 | | **组织架构** | 维持既有层级，将工具强塞给员工 | 最小可行人类组织（MVHO），工作流深度重构 | | **推进阻力** | 忽视基层恐慌，强制推行自上而下 | 心理安全感、包容异见、自然减员缓冲 | | **最终目标** | 局部自动化、降本裁员 | 赋能人类定义问题，实现全链路高确定性决策 | --- ### [18:51](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1131s) 心理安全感与求真文化：打完架再拥抱 在对谈的尾声，Bruce Smith 针对正在探索转型与创业的企业管理者提出了真诚的建议。他引用了史蒂夫·乔布斯（Steve Jobs）的经典名言：“你雇佣聪明人不是为了告诉他们该做什么，而是让他们来告诉你该做什么。” Bruce Smith 强调，领导者必须在组织内部建立起高度的“心理安全感”（Psychological Safety），让所有人都能坦然提出尖锐苛刻的问题，并得到真实客观的反馈。如果团队成员之间无法在业务现状、发展目标以及潜在障碍上保持绝对的坦诚，组织的发展方向必将陷入混乱。在 Voltava 内部，这种文化被形象地总结为：“有时我们需要痛痛快快地吵上一架，然后紧紧拥抱在一起，全速向前推进”（Sometimes we got to slug it out and then we hug it out and then we go）。只有建立在求真与信任之上的团队，才能在智能制造的新时代携手并进。 ## Notable quotes > [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 合作是成功者的秘诀，而非弱者的拐杖（Partnerships are the secret of the successful, not the crutch of the week） — Bruce Smith > [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 去芜存菁，辨别沙砾与真金（separate the glitter from the gold） — Bruce Smith > [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 当环境足够安全时，冲突就会高频发生（When the environment is safe, conflict is frequent） — Bruce Smith > [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 世上没有任何一辆汽车只有油门或只有刹车；我总是想开得更快（踩油门），Tom 总是想更稳健审慎（踩刹车），懂得在何种场景下如何配合施加力度，才是既安全又迅速抵达目的地的艺术。 — Bruce Smith > [02:09](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=129s) 我们可以忽视人工智能，但我保证，它绝不会忽视我们。 — Bruce Smith > [08:17](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=497s) 那就像别人在下国际象棋，而你还在原地玩西洋跳棋 — Tom Swartz > [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 我热爱员工并善用技术，绝不崇拜技术而利用员工。 — Bruce Smith > [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 你不会被 AI 取代，但可能会被使用 AI 的对手取代 — Bruce Smith > [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 行动的声音太大，以至于人们听不清你说了什么 — Bruce Smith > [10:11](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=611s) 大语言模型在公司毫无立足之地 — Voltava 员工 > [18:51](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1131s) 你雇佣聪明人不是为了告诉他们该做什么，而是让他们来告诉你该做什么。 — Steve Jobs（由 Bruce Smith 引用） > [18:51](https://www.youtube.com/watch?v=5RiFlJHUSyc&t=1131s) 有时我们需要痛痛快快地吵上一架，然后紧紧拥抱在一起，全速向前推进（Sometimes we got to slug it out and then we hug it out and then we go） — Bruce Smith ## Mentioned Voltava, Greggory Garrett, CGS Advisors, Bruce Smith, Tom Swartz, Kinetic, Steve Jobs（史蒂夫·乔布斯）, AI 大使机制 (AI Ambassadors), 公民笔记本计划 (Citizens' Laptops), 最小可行人类组织 (Minimum Viable Human Organization, MVHO), Ai4 ## Video description Greggory Garrett - CEO & Managing Director, CGS Advisors, LLC Bruce Smith - CEO, Voltava Tom Swartz - CTO, Voltava How does a Tier 1 manufacturing giant scale its global footprint when the old rules of productivity no longer apply? This fireside chat features the CEO and CTO of Voltava as they discuss their transition from a legacy manufacturing mindset to a modern Intelligence Strategy. We will explore how the fusion of human strategy and machine execution is helping them earn new work and transform their operations across the globe. The fireside chat will highlight the essential partnership between business and technology leaders and delve into Voltava’s unique bi-annual AI conferences that drive internal innovation. Attendees will see theory come to life through a real-world case study of organizational plasticity. Join this session to learn how to lead a firm where human high-order cognition and machine intelligence work in concert to create exponential value. [Source](https://www.youtube.com/watch?v=5RiFlJHUSyc)

[2:09] Bruce Smith: 合作是成功者的秘诀，而非弱者的拐杖（Partnerships are the secret of the successful, not the crutch of the week） [2:09] Bruce Smith: 去芜存菁，辨别沙砾与真金（separate the glitter from the gold） [2:09] Bruce Smith: 当环境足够安全时，冲突就会高频发生（When the environment is safe, conflict is frequent） [2:09] Bruce Smith: 世上没有任何一辆汽车只有油门或只有刹车；我总是想开得更快（踩油门），Tom 总是想更稳健审慎（踩刹车），懂得在何种场景下如何配合施加力度，才是既安全又迅速抵达目的地的艺术。 [2:09] Bruce Smith: 我们可以忽视人工智能，但我保证，它绝不会忽视我们。 [8:17] Tom Swartz: 那就像别人在下国际象棋，而你还在原地玩西洋跳棋 [10:11] Bruce Smith: 我热爱员工并善用技术，绝不崇拜技术而利用员工。 [10:11] Bruce Smith: 你不会被 AI 取代，但可能会被使用 AI 的对手取代 [10:11] Bruce Smith: 行动的声音太大，以至于人们听不清你说了什么 [10:11] Voltava 员工: 大语言模型在公司毫无立足之地 [18:51] Steve Jobs（由 Bruce Smith 引用）: 你雇佣聪明人不是为了告诉他们该做什么，而是让他们来告诉你该做什么。 [18:51] Bruce Smith: 有时我们需要痛痛快快地吵上一架，然后紧紧拥抱在一起，全速向前推进（Sometimes we got to slug it out and then we hug it out and then we go）

[2:08] Bruce Smith: Partnerships are the secret of the successful, not the crutch of the week. [2:08] Bruce Smith: We can ignore artificial intelligence, but I can guarantee you it won't ignore us. [9:57] Bruce Smith: I love people and I use technology. I don't love technology and use people. [9:57] Bruce Smith: What we do screams so loudly in everybody's ears, they can't hear what we say. [18:50] Bruce Smith: You don't hire smart people to tell them what to do. You hire smart people so they can tell you what to do. [18:50] Bruce Smith: Create the culture to where you can have real conversations with real people about real opportunities and move forward together in unison. And as we say internally, that means sometimes we got to slug it out and then we hug it out and then we go.

[0:04] Good afternoon everyone. I'm Greg Garrett. Uh  as was said, I'm the CEO and managing director [0:08] of CGS Adviserss. Uh just a quick moment to  understand this background. CGS for 15 years [0:14] has been working with uh leaders to try to help  them deal with disruption and bravely transform [0:20] their organizations to better compete in this ever  connecting world. And technologies around IoT and [0:26] uh industry 4.0 type technologies have been at  the forefront of that. And now obviously you're at [0:30] this conference especially artificial intelligence  and throughout the presentation you'll see these [0:35] QR codes. We're not trying to sell you anything.  If you just some of the thoughts and the [0:39] uh statistics that we're sharing are available  if you'd like to see the full reports. We we [0:44] interviewed several hundred sea level leaders  about AI transformation and and we're going to [0:49] lead that a little bit into the the conversation.  I've known these two gentlemen for a few years [0:54] now. Um, just a moment on Voltava. Uh, really  a global powerhouse, probably one that you may [1:00] not have heard of before. 21 manufacturing centers  around the world, over 5,000 employees, uh, really [1:07] making those operations hum every day and really  putting out a lot of very complex parts. Uh, [1:13] 30 million assemblies, so maybe not full vehicles,  but major components uh, of those working. and [1:20] and they're the premier assembler uh tier one  assembler for several industries like automotive [1:26] uh defense transformation these types of things  and with me on my left here as we said is Bruce [1:30] Smith he's not only the CEO but he's the chairman  and and owner I think he'll probably be one of the [1:35] few people at this conference that come from the  manufacturing industry that you can get that owner [1:40] operator experience from and that's why I asked  Bruce to join me to really challenge everyone [1:44] here on where AI needs to go not where it is going  where it needs to go to to drive value and on my [1:50] right is Tom Schwarz. Tom is uh got kind of the  difficult challenge to take the vision from Bruce [1:56] and make it somewhat of a reality at the same time  as making it scalable uh making it secure as the [2:03] CTO uh for Voltava. So let's just jump into it. Um  Bruce, maybe we can start with you a bit here. Uh [2:09] you know, as the owner operator of of the firm,  you and I have talked about technology. I think [2:14] we met maybe close to three years ago on the  topic of AI. fairly advanced for a for a CEO [2:19] of a manufacturing organization to be inviting in  people to to advise the organization on AI. And [2:24] you've always been sort of at the forefront  of technology. Um you get excited about it, [2:29] that's for certain. And and but you also have  a realistic view that you can't just uh chase [2:34] fools go gold as you like to say. What are some  of the hints that you have on on introducing [2:39] technologies like AI? And what kind of challenges  do you put in place at the same time as trying [2:43] to protect the organization maybe from some of  your excitement? Yeah. So, couple of things. Uh, [2:47] you're absolutely right. I'm excited about it.  But, uh, in in terms of being self-aware, I also [2:53] recognize that me being excited about it is both  a blessing and a curse for my organization. As you [2:58] know, when the boss gets excited about something,  it creates reactions throughout the organization. [3:04] And so, for me, it's really about the headline  of partnerships are the secret of the successful, [3:10] not the crutch of the week. And so Tom and I are  engaged as full partners to where I say to him, [3:15] "You're fully empowered to have veto rights over  whatever crazy idea I might bring." Because in [3:21] the end, as I'm sitting here being the visionary  of where I want us to go, how it ought to look, [3:26] how it ought to feel, you're the guy's got to put  the road map together to make it happen. Because [3:30] you're uniquely qualified to do that, whereas  I'm uniquely qualified to come up with the [3:35] next greatest idea. So I need you to separate  the glitter from the gold. And so he is fully [3:41] empowered to have 51% of the vote that when I  come in with something, he goes, "Thanks for [3:46] that idea. We're not doing it. The other one you  gave me, we will go do that." And then secondly, [3:51] in addition to he and I having a partnership to  where there's a good balance of power, there's [3:56] a lot of trust. And I'm a big believer that when  the environment is safe, conflict is frequent. And [4:01] so he has the opportunity to be invited into  disagreeing without being disagreeable so we [4:07] can get to the best idea cuz we want to do what's  right. I don't have any desire to be right. I just [4:12] want to do what's right. And then lastly, we've  got this interesting balance of no matter what the [4:17] topic is, I always want to go faster. He always  wants to go slower. So we often make the analogy, [4:23] I've never found one car that just had only a  brake in it or a car that only had a gas pedal in [4:28] it. And so in order to get somewhere, we've got to  have both a gas pedal and a brake. Knowing when to [4:34] apply which one at what time is really the art of  how you get from A to B safely and as quickly as [4:40] you can. Before we jump to Tom on how we actually  make some of that work, maybe just one follow-up [4:44] with you, Bruce, you're here at the AI conference  and you called me years ago. What is it about AI [4:49] that is got your attention from an owner uh you  know, kind of owner operator perspective? Yeah, [4:55] it's transformational. And so I can think about  uh throughout my career being an early adopter [5:01] and seeing technologies come on board whether  it was uh robots back in the late 80s and 90s or [5:08] uh when you think about for my industry when the  Japanese really got good at lean manufacturing and [5:13] you could see that starting to change the  industry. Here's another inflection point [5:17] uh laptop computers. I hate to age myself but  I remember when there was no such thing as a [5:22] laptop computer inside of a company. Here's  an opportunity to where work is going to be [5:28] changed permanently forever. And as I say to  my team, we can ignore artificial intelligence, [5:33] but I can guarantee you it won't ignore us. And so  figuring out how to embrace it and do it well and [5:40] do it quickly is going to be a key differentiator  for those who not only survive but thrive in the [5:45] future. Excellent. So Tom, you're on the receiving  end of uh of this vision, of this excitement. [5:50] Maybe you can just unveil a little bit uh  with a leader that's willing to partner, [5:54] give you 51%. How are you structuring things at  Voltava so that you can actually safely grab some [5:59] of these ideas and and explore them? Yeah, I  think one of the uh earlier accelerators was [6:04] uh Bruce took the fortitude to uh start another  company namely Kinetic that was non-automotive [6:10] and it really became a sandbox for us to to  work on innovation products, innovation ideas, [6:16] how we assemble, how we manufacture different  things and through that the course with an AI [6:22] we've utilized that to create uh agents and agents  talking to agents. We've got some CRM uh agents [6:29] that we've created to utilize that. Um but within  the walls of Voltaba, one of the unique things I [6:35] think we've done, we started about a year and a  half ago was identifying ambassadors. Uh we've got [6:40] approximately uh 15 ambassadors. Basically, what  they are is subject matter experts within their [6:46] departments that can help us identify areas of  of where we can use some use cases within AI. So, [6:53] they're their leaders. is they understand their  processes, their their everything that connects, [6:58] how it connects and then they come back to us  with working with an IT and the developers, [7:03] the engineers, and then we help them get those use  cases and put them into production. And then most [7:09] recently, one of the things that uh we've started  to do is uh we've identified about 20 different [7:14] people that just have a real desire to get into  the AI, whether it's no coding, low coding, [7:19] creating agents, and obviously with governance and  everything that goes along with it. Having people [7:24] be able to do some of that those things on our  network is is problematic. So we've actually we're [7:30] calling citizens laptops that we're issuing out  to you know 20ome people that allow them to hey if [7:36] you want to work on coding no coding or creating  agents to keep it off the network we'll help with [7:41] the training and things like that so then that  they can get used to it and then we if we find [7:46] uh some opportunities we'll bring them onto the  network. So, I really celebrate that two-prong [7:50] approach uh for disruption to the business model,  for disruption to uh differentiators. You've got [7:56] kinetic uh standalone business unit measured  in a different way probably on some different [8:00] metrics and this uh 18 15 18 person uh cross  functional team really working through likely [8:08] uh being able to drive from organizational design  through agentic design and that eventually gets [8:13] back to the tech. It keeps things in lock step  so not one of them gets ahead of the next one. [8:17] I think it's it's it's quite useful. Maybe on a  slightly different uh note, you and I chatted a [8:23] little bit about uh the the importance of data  and how there's likely some data vendors that [8:29] are out in the expo hall that will suggest that  you need to spend, you know, millions of dollars [8:33] and several years to transform, especially the  way you guys have grown through acquisition and [8:37] bolting companies together, that you're going  to have to a major cleanup. And you likely do, [8:41] but do you have to wait for it? And if not, what  are some of the hints that you you may share on [8:46] not waiting for data to be ready? Yeah, the old  data dilemma whether you should wait and you'll [8:51] hear it many times, but in my opinion,  absolutely not. There's um if if you do, [8:56] you're playing uh checkers where everybody else  is playing chess. You really have a lot of swim [9:02] lanes that you need to start developing your  AI strategy, roadmap, you know, governance, [9:07] um education, uh getting buyin from people, um  etc. So within all those seven I think we've got [9:14] seven swim lanes we've identified but within those  swim lanes you got to make sure that you're contin [9:19] continuously pushing those envelopes and getting  your AI maturity level ready. Sure data is the the [9:25] foundation. There's no doubt about it. It is the  the rock of AI. But um but if you wait for it to [9:31] be ready it likely will change and it'll continue  to change. It'll have agents that are changing [9:35] it for you. So it's um it it's in parallel with  everything else that needs to happen. That's good. [9:42] And and really thinking of that as a reasoning  ceiling, you know, how can you move synthetic [9:46] data? How can you move different types of sources  of information to start practicing as you said, [9:51] I think is absolutely a key and something we're  seeing, you know, in Voltava, but but also beyond. [9:57] I one of this this this whole premise and you  heard in the in the setup of the importance of not [10:03] only the machines or the artificial intelligence  but also the humans piece of it. So I just want [10:07] to introduce this concept to the to the group here  and then be able to maybe ask you a question about [10:11] it Bruce is when we've been looking at really  what makes uh artificial intelligence work. We're [10:18] not studying we've got a general belief that the  technology is going to get there. In fact, I think [10:23] if you walk the expo halls, it's advancing at a  rate that most organizations can't keep up with. [10:28] The real secret is making that technology work in  parallel with the humans inside the organization. [10:33] Until we get to uh you know AGI, I think that  we're going to have to have a blended uh uh [10:39] program. And so it's this is not an additive type  of uh technology. It's not human intelligence plus [10:44] machine intelligence. This is a multiplicative.  And it's about getting that density right, [10:48] figuring out where humans should be and what their  roles are and where machines should be and what [10:53] its roles uh should look like and getting that  density right can be exponential. So, you know, [10:58] to that end, Bruce, you've prided as you've bolted  these companies together and built this, you know, [11:02] major global powerhouse. You've really had a  lot of pride in the culture of the organization, [11:07] changing the culture, getting people to pull  in one line. So, just speak with your energy [11:11] to bring the technology in. How do you protect  the humans? What is the role of the human? and [11:15] how do you really seek uh advancing the human in  this whole model? Yeah, I I think it's important [11:20] to be clear about what you care about and so  uh one of the phrases I'll often use is uh I [11:27] love people and I use technology. I don't love  technology and use people. And your folks can [11:33] tell what's most important to you. And so for  us, it's really about creating this glide path [11:39] to where it's not either or because we often  say, you know, we're not going to lose our [11:43] company or you're not going to lose your job to  AI, but we may lose our company and we may lose [11:48] our jobs as someone else who's using AI. So, how  do we create this vin diagram to where there's [11:53] this overlap between what you do and what AI does  as opposed to being just a replacement feature? [11:59] And to the extent we can continue to implement  all the agents and and productivity improvements [12:05] that we're driving and allow natural attrition  to take care of where we don't backfill people. [12:11] We've got something that's pretty harmonious and  isn't creating this huge fear around uh hey if [12:17] I embrace AI they're going to let me go the week  after. And so we're really deliberate about saying [12:22] we're doing this with and for our people and not  doing it to our people. And ultimately our actions [12:29] which the old saying what we do scream so loudly  in everybody's ears they can't hear what we say. [12:35] We truly have been practicing that and continue  to be committed to practicing that so that we [12:40] don't create unnecessary fear. And then the last  thing I'll add is you I had a boss years ago who [12:46] would say people are going to talk. It's your job  to help them to have good things to talk about. [12:51] uh we talk about AI and so uh otherwise you're  leaving it to the press and what people read and [12:57] hear to fill in their own blanks about how big  of a monster AI is or isn't. And so we openly [13:04] talk about it. Uh Tom had someone the other day  who told him large language models don't have any [13:09] place in the company. What did Tom do? He reached  out and said, "Hey, why'd you say that? Let's talk [13:14] about it." U and didn't there was no retribution  or punishment. that person is still gainfully [13:20] employed and has a role at our company, but they  had the courage to raise their hand and say they [13:25] disagree and we leaned into that. And so how you  behave really tells people how you're approaching [13:31] this and what you care about. And again, we're  doing it with our folks and not to our people. [13:36] You know, I think for those of you that aren't  from the manufacturing sector industry, listening [13:41] to a CEO talk about we talk about AI, that's it's  it's a progressive uh it's a progressive look, [13:48] especially when you get uh bluecollar labor in and  and uh uh unions involved and whatnot. So, this is [13:55] a kind of a progressive lean in to really drive  it out. And I'd like to put a challenge out if [14:00] anybody's from the from the conference. I'd love  to be seeing a a CHRO track uh started in this [14:06] conference because the reality is the human piece,  the tech is going to get there. The the humans [14:11] really being part of this is going to really be a  important role. We like to talk and in in some of [14:16] these papers that we've written about the minimum  viable human organization and that doesn't mean [14:20] like get rid of the humans. It means we only  need the humans that can do the things that [14:24] humans have to do. And how do we raise that skill  level up? And you can see the quote at the bottom [14:29] of there that problem articulation and framing is  likely going to be the number one competency in [14:36] the agentic era for leaders. That's a pretty, you  know, that's a big shift for a lot of companies [14:42] that says we need people that are can identify  problems and frame them verse do work. And [14:48] uh it's going to take all of us kind of pulling  in that same direction. So maybe back to you Tom [14:52] for to start kind of closing in here. you've got a  likely some different types of people in the room, [14:58] partners, vendors, manu maybe some competitors,  other manufacturer tiers, whatnot. As you're [15:03] starting to think through needing to get this  stuff done and uh taking some of Bruce's vision [15:08] and surrounding people in your organization with  this technology to enable them, what are some of [15:12] the edge things that you're trying to get to that  you think you may need some help with? What what [15:15] kind of challenges can you give to the the crew  here to to come find you in the hallway after say, [15:20] "I think I can help you with this or help you with  that?" Yeah, I [clears throat] um as far as like [15:24] vendors go, I mean, we've got a lot of suppliers.  I've got them lined up at the office all the time. [15:29] They're on I mean, we've get plenty of them and  I I enjoy working with some of them. But really, [15:34] what we're looking to do is find those use cases  in the manufacturing sector, whether on the shop [15:40] floor, back office, um where we can actually put  them in production. So any anybody that can help [15:46] us out with understanding whether whether it's  supply chain or finance or you know where you've [15:52] been successful at um and and being boots on the  ground maybe some engineers or developers etc. So [15:58] we can always definitely use that. Um as far as  competitors and other people in the manufacturer [16:03] like we talked about it's time to take silos  down. We have to have these conversations. We [16:08] have to have even if it's combative we we need to  hear everybody's voices. I mean it happens in the [16:13] backyards with data centers. I mean these things  are happening. We understand they're happening. [16:17] Let's collectively let's talk about that and  how we can help each other out. Um I sit on [16:21] the council with one of our uh our customers  and and we have all of our competitors talking [16:26] talking within amongst ourselves because there  are how are we going to navigate through these [16:30] challenging times of governance and and data  wiping things out and agents going in and going [16:36] into their data and things like that. So let's  have those conversations. And then uh as far as um [16:41] anybody you know customers and potential customers  it's let's let's again let's work together around [16:47] AI it's better to work with with each other  sharing the data sharing the use cases and [16:54] um and and let's let's get through this together  and it's a it's a challenging time for all of us. [17:00] So an exciting time too. Perfect. Tom the uh  again if you're not super familiar with the [17:06] manufacturing industry especially the automotive  industry which is where a lot of these two guys [17:10] spend a lot of their time it's a it's a bit of a  fragmented market years ago uh the verticalization [17:17] of the industry kind of split across and you've  got sourcing of goods from many different lands [17:22] and a lot of handoffs and if you think about the  promise of artificial intelligence sometimes you [17:27] actually need to look pull back the lens a little  bit to be able to remove steps so partnership [17:32] working on challenges together is really going  to be something that is likely a big piece of the [17:36] future for the manufacturing sector. So, as we as  we wrap up here and get the uh the day continued, [17:43] uh I'd like to just kind of out a couple of things  that these two gentlemen have put forward is that [17:48] as Bruce has pushed forward uh to towards you. A  lot of this uh the success in in many sectors, but [17:54] manufacturing absolutely included is it's probably  not about the fundamental shift of technology. [18:00] uh it's probably more of the fun fundamental  shift of the executive mindset that needs to be [18:05] uh tuned towards and if you if you're running  AI transformation programs and you don't have [18:11] a a organizational redesign component if you don't  have a a workflow uh redesign if you don't have a [18:18] as as Bruce was suggesting a model where you bring  the people in that have the concern that are going [18:23] to act as the inert force you probably have some  pretty major uh aspects of your of your program [18:29] that are missing. And uh I I'll remind you back  to that equation that we put up there. It's likely [18:34] really less about an AI strategy and more about an  intelligence strategy in order for for you to get [18:39] there. And the and the main goal uh of your people  is going to be about getting decisions making for [18:45] certain. So uh I'm going to flip to this slide and  and maybe just ask one more question over here to [18:51] my my friend Bruce is we've we've collected  a lot of information. We you can again scan [18:56] that and see a lot of a lot of the uh the longer  form versions of this. But Bruce, if you can leave [19:02] the room a little bit with a challenge as a as a  guy who's built businesses uh driven businesses, [19:08] you got a new industry out here, an AI industry,  what kind of hints would you give to them as as [19:13] a business builder uh to see to put them on a  plateau to be able to have some success? Yeah, [19:18] I would start with a quote by Steve Jobs who said,  "You don't you don't hire smart people to tell [19:23] them what to do. You hire smart people so they can  tell you what to do." So, create a system to where [19:28] you really do have uh psychological safety and  where you get a chance to ask hard questions and [19:36] get honest answers to them. Because if we're not  being truthful with one another about what's going [19:41] on and what we're trying to do and what things  can enable us or disable us from doing that, [19:46] if we're not being truthful with one another, I  don't know where we're headed, but I don't like [19:50] the destination. And so create the culture to  where you can have real conversations with real [19:55] people about real opportunities and move forward  together in unison. And as we say internally, [20:00] that means sometimes we got to slug it out  and then we hug it out and then we go. And so, [20:06] uh, I'd leave the audience with  that. That's perfect. Bruce, Tom, [20:09] thank you for being here. Everybody, enjoy the  rest of your day. Thank [applause] you. Thank you.