Dice Finds 75% of AI Talent Eye Exits as Automated Hiring Fails

Finding the Right Talent in the AI Era With DHI Group, Inc — Ai4

Ai4

Over-reliance on automated screening bots alienates in-demand AI professionals who distrust the technology in recruitment, making transparent human evaluation essential to retaining talent.

AI-assisted summary of Finding the Right Talent in the AI Era With DHI Group, Inc — original source. Check the original for full context.

Source published: 2025-08-24

Detailed notes

## Background Published on YouTube by the Ai4 channel on August 24, 2025, this presentation features Brooke Lorenz, Director of Market Research and Insights at DHI Group, Inc. According to the description, the session delivers research-backed insights and Dice hiring data to help C-suite leaders attract, evaluate, and retain top AI talent while avoiding common strategic pitfalls.

## TL;DR Brooke Lorenz demonstrates that the primary bottleneck in enterprise AI transformation is human talent rather than hardware or data infrastructure. She outlines the three tiers of the AI talent ecosystem—Builders, Orchestrators, and Enhanced Professionals—and examines hiring trends expanding into traditional industries and non-coastal hubs. The session critically analyzes the flaws of fully automated recruitment funnels, where algorithmic screening alienates candidates and clashes with widespread applicant AI use. Ultimately, Lorenz advises leaders to adopt transparent hybrid hiring models, grant engineers strategic autonomy, and prioritize cultural alignment to retain highly mobile AI practitioners.

## Atmosphere & style The presentation maintains an engaging, pragmatic, and data-driven tone, kicked off with self-deprecating humor as Lorenz acknowledges standing between a Las Vegas audience and happy hour. It feels like an empathetic reality check for corporate leadership, highlighting the ironies of modern hiring where automated applicant bots clash with automated screening filters. The delivery is conversational and clear, grounding high-level executive strategy in candid candidate sentiments and actionable labor market insights.

## Key takeaways - [00:04](https://www.youtube.com/watch?v=Qhr1zALLZao&t=4s) The primary bottleneck in enterprise AI transformation is human talent rather than technology or capital, with nearly 50% of investing firms lacking the in-house expertise to execute their strategy. - [00:04](https://www.youtube.com/watch?v=Qhr1zALLZao&t=4s) AI competencies are transitioning into mainstream technical requirements, now appearing in nearly 40% of all US tech job postings. - [03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) The AI talent ecosystem is divided into three distinct tiers: Builders (core ML researchers), Orchestrators (technical and product translators), and Enhanced Professionals (workers augmenting existing workflows with AI). - [03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) Orchestrators represent the fastest-growing hiring category for enterprise AI deployment, while Enhanced Professionals constitute the largest untapped pool for internal upskilling. - [05:49](https://www.youtube.com/watch?v=Qhr1zALLZao&t=349s) AI hiring demand is expanding geographically beyond traditional tech hubs, with Texas outpacing California in year-over-year growth and Minnesota recording a 110% annual increase. - [05:49](https://www.youtube.com/watch?v=Qhr1zALLZao&t=349s) Traditional industries such as manufacturing, healthcare, aerospace, and consulting are actively recruiting AI practitioners to modernize enterprise operations. - [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) Automated recruitment screening often creates distrust, with 68% of tech professionals distrusting fully automated hiring pipelines and 46% preferring to opt out of automated resume filtering. - [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) While 82% of applicants use generative AI tools to draft job applications, rigid keyword filtering misses adaptable talent with high learning agility. - [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) AI professionals exhibit high mobility, with roughly 75% planning to change employers within 12 months and 60% expressing confidence in landing favorable new positions. - [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) Non-tech enterprises hold significant recruiting appeal, as 71% of AI practitioners are open to working outside of Big Tech in exchange for technical ownership and measurable impact. - [19:11](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1151s) Leaders must articulate a distinct value proposition and technical autonomy, using internal discovery roundtables to uncover why current engineers stay engaged. - [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) Sustainable AI talent acquisition relies on a hybrid model that uses AI for operational efficiency while preserving human connection, transparency, and cultural alignment.

## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=Qhr1zALLZao&t=4s) The Real Bottleneck in AI Transformation

Brooke Lorenz opens her presentation with a humorous nod to the audience in Las Vegas, acknowledging that as an introverted researcher, she knows she stands directly between the attendees and happy hour. Clarifying that Dice is a tech career marketplace rather than a manufacturer of craps dice, Lorenz outlines the platform's role in connecting employers with skilled tech professionals while guiding candidates through their career journeys. She presents a scenario familiar to almost every corporate boardroom: leadership declaring the urgent need to invest millions into artificial intelligence, pouring capital into advanced data pipelines, computing models, and infrastructure, while completely overlooking the human talent necessary to bring those systems to life.

``` AI TRANSFORMATION BOTTLENECK +-------------------------------------------------------------+ | Heavy Capital Investment --> Data, Models, Infrastructure | +-------------------------------------------------------------+ | v (The Talent Disconnect) +-------------------------------------------------------------+ | 40% of US Tech Job Postings Require AI Skills | | Nearly 50% of Investing Firms Lack In-House AI Expertise | +-------------------------------------------------------------+ | v +-------------------------------------------------------------+ | REAL BOTTLENECK: Attracting, Evaluating, & Retaining Humans| +-------------------------------------------------------------+ ```

This disconnect creates a massive organizational vulnerability. Citing research from Bain & Company, Lorenz notes that nearly half of companies investing heavily in AI technology stacks admit they lack the internal expertise required to deploy and maintain them. Meanwhile, AI skills have transitioned from a niche specialty to a mainstream requirement found in nearly 40% of all tech job postings across the United States. Demand has outpaced the supply of experienced practitioners, leaving organizations staffed inadequately for the systems they have purchased. As Lorenz emphasizes, the true bottleneck of AI transformation is not the technology itself, but the people behind it.

- According to Bain & Company, nearly 50% of enterprises investing millions in AI infrastructure admit they lack the in-house talent to execute their strategy. - AI competencies are now explicitly required in nearly 40% of all US tech job postings. - High-performing AI professionals are highly mobile, selective, and scrutinize how prospective employers communicate and demonstrate technical maturity.

---

[03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) The Three Tiers of the AI Talent Ecosystem

Many business leaders mistakenly envision AI talent as a single archetype, often captivated by sensational headlines such as Meta extending nine-figure compensation packages to poach top OpenAI researchers. Lorenz explains that the talent ecosystem is far more nuanced, dividing the landscape into three distinct categories: Builders, Orchestrators, and Enhanced Professionals.

``` THE AI TALENT ECOSYSTEM

[ BUILDERS ] [ ORCHESTRATORS ] [ ENHANCED PROF. ] - Core AI Researchers - AI Product/Program - Software Engineers - ML Engineers Managers - Solutions Architects - Data Scientists - Automation Leads - Tech Marketers - Deep Technical Focus - Business Translators - Copilot/Claude Users ---------------------- --------------------- ---------------------- Smallest talent pool, Fastest-growing tier; Largest untapped pool; intense competition & bridges model tech & ideal for upskilling extended hiring cycle commercial outcomes & internal mobility ```

Builders comprise core AI researchers, machine learning engineers, and data scientists with deep technical foundations. This represents the smallest talent pool, characterized by fierce hiring competition, heavy recruiter outreach, and protracted decision cycles as candidates carefully evaluate every technical environment and culture. In contrast, Orchestrators—including AI product managers, technical program managers, automation specialists, and data analysts—bridge the divide between commercial objectives and model capabilities. Lorenz highlights Orchestrators as the fastest-growing segment and a prime hiring sweet spot for enterprises looking to operationalize AI.

Finally, Enhanced Professionals constitute the largest untapped talent pool. This group includes existing software engineers, solution architects, and even marketers who actively augment their daily workflows with tools like GitHub Copilot and Anthropic's Claude. Despite being primed for internal mobility and upskilling, these professionals are frequently overlooked by corporate recruiters simply because they lack explicit AI keywords in their official job titles.

---

[05:49](https://www.youtube.com/watch?v=Qhr1zALLZao&t=349s) Geographic Shifts and Industry Demand Patterns

Analyzing US tech hiring trends using Lightcast labor market intelligence, Lorenz illustrates where AI hiring demand is concentrating and expanding. While California continues to lead the nation in absolute job posting volume, other regions are seeing faster percentage growth. Texas has outpaced California in year-over-year AI job posting growth, and Minnesota has emerged as a rapid-growth market with a 110% year-over-year increase.

``` AI HIRING EXPANSION MAP

Hiring demand is also expanding outside traditional Big Tech firms. Heavy industries such as manufacturing, healthcare, and aerospace are aggressively recruiting technical specialists to modernize operations, competing directly with Silicon Valley giants. Furthermore, professional services and consulting firms—including Deloitte, Accenture, and PwC—maintain a steady pipeline of engineers who have built enterprise-scale AI solutions across diverse sectors. These practitioners are increasingly receptive to in-house corporate roles that offer direct ownership and measurable business impact.

- California retains the highest gross volume of AI job openings in the US. - Texas surpasses California in year-over-year growth rates for AI positions. - Minnesota recorded a 110% year-over-year increase in AI-related job postings. - Traditional industries like aerospace and manufacturing represent prime destinations for AI talent seeking stable, high-impact enterprise initiatives.

---

[08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) Breaking the Cycle of Broken AI Recruitment

Lorenz warns that many organizations undermine their talent acquisition efforts through impersonal and automated screening processes. Candidates frequently encounter situations where a company claims to be an "AI-first" innovator, yet subjects applicants to disconnected resume parsing bots or issues automated rejections within ten minutes of submission. In a Dice survey assessing sentiment toward AI-driven recruitment, one candidate remarked: "It doesn't look cutting edge. It looks foolish. It shows you don't understand the technology. Why would I want to work for you?"

``` THE AUTOMATED HIRING PARADOX

Lorenz highlights two primary missteps in technical recruitment:

1. **Optimizing for Job Descriptions Over Human Potential**: Rigid keyword filtering causes recruiters to miss adaptable talent. For instance, prompt engineering was treated as a distinct specialty just a couple of years ago, but has quickly evolved into a baseline skill. Additionally, aggressive 30-day hiring sprints routinely fail with senior AI talent, who require adequate time to evaluate an organization's mission, team dynamics, and technical stack. Copy-pasting boilerplate job descriptions signals a fundamental misunderstanding of the role, whereas effective organizations evaluate candidates for learning agility, establish close alignment between recruiters and engineering managers, and craft job postings that reflect actual team culture.

2. **Over-Reliance on Automated Screening Tools**: Survey data reveals that 68% of tech professionals distrust fully AI-driven hiring workflows, and 46% would opt out of automated resume screening if given the choice. Ironically, the practitioners who understand AI best exhibit the greatest skepticism toward its use in hiring. This dynamic is compounded by candidate behavior: 82% of tech workers report using generative AI tools to craft applications and resumes, with 51% unconcerned about recruiter detection. To pierce through AI-generated noise, one recruiter described asking applicants five targeted, bespoke questions, feeding the responses into ChatGPT to identify and filter out generic or repetitive answers, and advancing candidates whose submissions displayed genuine domain comprehension.

- 68% of tech professionals distrust completely automated, AI-driven hiring processes. - 46% of candidates would choose to opt out of automated AI resume screening if offered an alternative. - 82% of tech professionals use AI tools during their job application and resume preparation process. - 51% of tech applicants have no concern regarding whether employers detect AI assistance in their submissions.

---

[13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) Attracting and Retaining Mobile AI Talent

Transparency serves as the foundation for attracting high-caliber AI practitioners. Technical candidates look for transparent salary bands, clear outlines of the interview process, and realistic technical requirements—noting that unrealistic listings asking for 15 years of ChatGPT experience destroy employer credibility. Survey respondents highlighted that automated filters fail because "AI is configured to look for specific things," ignoring the reality that "careers in tech are rarely linear and that complexity gets lost in automated screening." Beyond base compensation, top talent evaluates roles based on strategic influence, technical autonomy, and whether they are creating functional, production-grade systems rather than disposable prototypes.

``` AI TALENT FLIGHT RISK

[ AI PROFESSIONALS ] [ GENERAL TECH WORKERS ] =============================================================== Confident in finding a favorable role: [ 60% ] [ 36% ] --------------------------------------------------------------- Planning to switch employers within 12 months: [ ~75% ] [ 65% ] =============================================================== RETENTION DRIVERS: * Demonstrable business impact & strategic autonomy * Dedicated learning budgets to keep pace with tooling shifts * Culture valuing technical curiosity, ethics, and influence ```

Retention poses an acute challenge because AI professionals recognize their market leverage. In Dice's survey, nearly six in ten AI specialists expressed confidence in securing a favorable new role, compared to only 36% of general tech professionals. Furthermore, almost 75% of AI practitioners reported plans to change employers within the next twelve months, compared to 65% of the broader tech workforce. Lorenz cautions employers to monitor early warning indicators of disengagement, such as engineers withdrawing from internal technical discussions, resisting new initiatives, or stepping back from team collaboration.

To maintain long-term commitment, leaders must provide continuous learning budgets, meaningful technical challenges, and a culture centered on curiosity and ethics. Lorenz concludes with an encouraging insight for non-tech enterprises: while 29% of surveyed AI professionals prefer Big Tech environments, 71% are open to other business types. Traditional industries, such as manufacturing, hold distinct advantages in predictive maintenance, automation, and long-term business stability that Big Tech cannot always provide.

- 60% of AI professionals are confident in their ability to land a favorable new position, compared to 36% among the wider tech workforce. - Approximately 75% of AI professionals plan to change employers within the next year, compared to 65% across general tech roles. - 71% of AI talent is receptive to employment outside traditional Big Tech firms.

[19:11](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1151s) Finding Your Edge and Uncovering Hidden Strengths

Regardless of the industry an organization operates in, leaders must get crystal clear on their unique value proposition and organizational edge. Competing for elite tech talent requires establishing a direct connection between day-to-day work and the broader organizational mission, while highlighting opportunities for real technical ownership. For leaders unsure of what their organization's specific strengths are, the best path forward is internal discovery: engage directly with the existing tech workforce by hosting roundtables to ask candidly what keeps them around and engaged.

- **Clarify the organizational edge:** Tie engineering roles directly to meaningful mission and impact. - **Highlight technical ownership:** Show candidates they will have autonomy and responsibility over their projects. - **Uncover hidden strengths:** Conduct internal roundtables with current technical staff to learn why they choose to stay.

---

[19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) Core Takeaways: Why AI Strategy Starts with People, Trust, and Hybrid Hiring

Translating these insights into practical execution comes down to a fundamental shift in perspective: "AI strategy doesn't start with models. It starts with people." The way an organization hires directly shapes the team, which in turn determines the success of every downstream AI initiative. Because top AI talent has abundant options and understands their market worth, opportunities must distinctly stand out to win them over.

Furthermore, while organizations feel immense pressure to hire as quickly as possible, trust and transparency ultimately win over sheer speed in fiercely competitive talent markets. Cultivating genuine relationships throughout the hiring journey yields far better long-term outcomes than rushing candidates through an impersonal pipeline. Balancing this requires a hybrid hiring model. Given the sheer volume of applicants, relying purely on manual human review is unsustainable; however, turning the process entirely over to automated AI screening creates alienating candidate experiences. Organizations must leverage AI tools for efficiency while keeping authentic human connection alive at critical touchpoints. Finally, cultural alignment is paramount for long-term retention—technical tools and frameworks can be taught on the job, but culture fit cannot. Ultimately, professionals stay where they feel seen, challenged, and respected.

- **People over models:** AI strategy succeeds or fails based on the quality of talent acquisition and team composition. - **Transparency beats pure speed:** Building relationships and maintaining open communication during recruitment outperforms transactional, rushed hiring. - **The hybrid recruiting imperative:** Combine AI-driven efficiency for handling application volumes with human touchpoints to preserve empathy and candidate engagement. - **Culture drives retention:** Prioritize cultural alignment alongside technical skills, ensuring employees feel valued, challenged, and respected.

## Notable quotes > [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) It doesn't look cutting edge. It looks foolish. It shows you don't understand the technology. Why would I want to work for you? — Survey Respondent

> [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) AI is configured to look for specific things — Survey Respondent

> [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) careers in tech are rarely linear and that complexity gets lost in automated screening. — Survey Respondent

> [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) AI strategy doesn't start with models. It starts with people. — Brooke Lorenz

## Mentioned Brooke Lorenz, DHI Group, Inc, Dice, Ai4, Bain & Company, Meta, OpenAI, Anthropic, Claude, GitHub Copilot, ChatGPT, Lightcast, Deloitte, Accenture, PwC

## Video description Brooke Lorenz, Director, Market Research and Insights, DHI Group, Inc.

Abstracts

For C-suite leaders beginning their AI journey, this session offers research-backed insights into building high-performing AI teams in a fiercely competitive talent market. Drawing on new survey data from tech professionals and hiring trends from Dice, you’ll learn how to attract, evaluate, and retain top AI talent by building trust, while avoiding the strategic missteps that cause organizations to lose them.

[Source](https://www.youtube.com/watch?v=Qhr1zALLZao)

中文简报

Dice调查：自动化招聘失效，75%的AI人才考虑离职过度依赖自动化筛选机器人疏远了紧俏的AI专业人才，他们对招聘中的此类技术缺乏信任，这使得透明的人工评估成为留住人才的关键。 Brooke Lorenz指出，企业AI转型的首要瓶颈在于人才，而非硬件或数据基础设施。她梳理了AI人才生态的三个层级——构建者（Builders）、协调者（Orchestrators）和增强型专业人员（Enhanced Professionals），并探讨了招聘需求向传统行业及非沿海枢纽扩散的趋势。本次分享深入剖析了全自动化招聘漏斗的弊端：算法筛选不仅疏远了候选人，也与求职者广泛使用AI的现状相冲突。最终，Lorenz建议管理者采用透明的混合招聘模式，赋予工程师战略自主权，并重视文化契合度，以留住流动性极高的AI从业人员。企业 AI 转型的核心瓶颈在于人才而非技术或资本。贝恩公司数据显示近 50% 投资 AI 的企业缺乏内部专业人才，而全美近 40% 的科技职位已将 AI 技能列为硬性要求。 (0:04) AI 人才生态可划分为构建者（Builders）、协调者（Orchestrators）和增强型专业人员（Enhanced Professionals）；其中协调者是落地部署增长最快的岗位，而增强型人员则是最大的内部技能提升潜力池。 (3:38) 招聘需求正在向传统科技枢纽之外扩散，得克萨斯州的职位年增长率已超越加州，明尼苏达州激增 110%；同时制造业、医疗、航天与咨询等传统行业正加速招揽 AI 人才推动现代化。 (5:49) 过度依赖自动化招聘工具引发信任危机，68% 的技术人才不信任全自动招聘流程，46% 希望退出简历自动过滤；而 82% 的求职者已在借助生成式 AI 编写求职材料，僵化的关键词筛选极易错失高敏捷度人才。 (8:07) AI 专业人才具有极高的市场流动性，近 75% 计划在未来一年内更换雇主，60% 对找到理想职位充满信心；但同时有 71% 的人对传统科技巨头（Big Tech）以外的行业机会持开放态度。 (13:41)

Reading summary

Brooke Lorenz of DHI Group argues that the decisive bottleneck in enterprise artificial intelligence adoption is human talent rather than computing infrastructure, requiring organizations to restructure their hiring and retention strategies for an increasingly mobile workforce. The Capital Disconnect and Talent Ecosystem Tiers (0:04) Lorenz observes that while corporate leadership frequently commits substantial capital to AI infrastructure, data pipelines, and computing models, execution falters due to a lack of human expertise. Citing research from Bain & Company, she notes that nearly 50% of enterprises investing heavily in AI technology stacks admit they lack internal talent to deploy and maintain them, even as AI proficiencies have become explicit requirements in approximately 40% of US tech job postings. To address this gap, Lorenz rejects viewing AI talent as a single archetype, segmenting the workforce into three operational tiers. 'Builders'—machine learning engineers and core researchers—constitute the smallest and most fiercely contested segment. 'Orchestrators'—including AI product managers, automation leads, and data analysts who translate business goals into technical requirements—represent the fastest-growing hiring tier. Finally, 'Enhanced Professionals'—such as software engineers and technical marketers using tools like GitHub Copilot or Anthropic's Claude—comprise the largest untapped talent pool, though corporate recruiters frequently overlook them because their titles lack specific AI keywords. Geographic Diversification and Cross-Industry Demand (5:49) Drawing on Lightcast labor market intelligence, Lorenz demonstrates that hiring demand is expanding beyond traditional technology hubs. While California maintains the largest raw volume of AI job postings, Texas has outpaced California in year-over-year job growth, and Minnesota has recorded a 110% year-over-year increase in AI-related postings. Concurrently, hiring pressure is diversifying across heavy industries, with aerospace, healthcare, and manufacturing competing directly against Silicon Valley firms. Professional services firms such as Deloitte, Accenture, and PwC also train large pools of enterprise engineers who are receptive to in-house corporate roles offering direct ownership. According to Dice survey data, 71% of AI professionals are open to working outside traditional Big Tech, giving non-tech enterprises an advantage if they can offer stability and impactful operational projects like predictive maintenance. Automated Recruitment Flaws and Candidate Distrust (8:07) Lorenz warns that impersonal, automated screening pipelines actively damage employer credibility. Survey findings indicate that 68% of tech professionals distrust fully automated hiring processes, and 46% would choose to opt out of automated resume filtering if given an alternative. Rigid keyword parsers frequently penalize non-linear career paths and fail to capture learning agility, even as technical requirements shift rapidly—illustrated by prompt engineering moving from a standalone specialization to a baseline skill within two years. Recruiting complexity is further elevated by applicant tooling: 82% of tech professionals report using generative AI to prepare applications, with 51% unconcerned about employer detection. Lorenz advises adopting a hybrid recruitment model rather than fully automated screening sprints. Organizations can employ AI tools to manage high application volumes, but they must maintain human touchpoints and deploy targeted domain questions to evaluate genuine comprehension and cultural alignment. Heightened Mobility and Retention Strategies (13:41) Retention represents an acute operational risk because AI practitioners recognize their labor market leverage. In Dice's survey, 60% of AI specialists expressed confidence in securing a favorable new position—compared to 36% of general tech workers—and roughly 75% reported plans to change employers within 12 months, versus 65% across the broader tech workforce. To counter turnover, Lorenz urges employers to monitor disengagement indicators, such as withdrawal from technical collaboration, and establish clear organizational value propositions. Beyond competitive and transparent salary bands, top practitioners evaluate prospective roles based on technical autonomy, strategic influence, dedicated learning budgets, and the opportunity to build production-grade systems rather than disposable prototypes. She recommends holding internal roundtables with existing technical staff to identify the cultural and project attributes that foster long-term loyalty. Lorenz concludes that enterprise AI strategy succeeds or fails based on team composition rather than algorithmic complexity. While organizations face pressure to hire rapidly, establishing relationship-driven transparency and cultural alignment ultimately outperforms automated, transactional recruiting in securing high-demand technical talent.

精炼阅读摘要

Dice 研究员 Brooke Lorenz 指出，企业人工智能转型的关键瓶颈在于人才的获取、评估与留存，而非技术硬件或算法模型。企业若要弥合战略雄心与落地执行之间的断层，必须纠正过度依赖自动化筛选的招聘偏差，转向透明且注重人际沟通的混合招聘模式。资本投入与人才断层的现实困境 (0:04) Lorenz 强调，许多企业董事会投入数百万美元采购算力、数据管道和高级计算模型，却往往忽视了实际部署这些系统所必需的技术团队，形成了显著的组织脆弱性。引用贝恩公司（Bain & Company）的研究数据，近 50% 在人工智能基础设施上大量投资的企业承认，内部缺乏执行与维护相关战略所需的技术人才。与此同时，人工智能技能已从边缘小众专长转变为通用要求，目前全美约 40% 的技术招聘职位已明确要求具备 AI 相关能力。由于高水平专业人员供不应求，大量企业面临系统已购置到位却无人落地的困局。顶尖的 AI 从业人员流动性极强且选择审慎，在求职时会严格审视潜在雇主的技术成熟度与沟通方式。Lorenz 据此断言，AI 转型的真正阻碍并非技术本体，而是技术背后的人才断档。 AI人才生态的三层细分架构 (3:38) 许多企业管理者将 AI 人才误判为单一群体，过度受媒体关于顶尖研究员天价薪酬报道的影响。Lorenz 将 AI 人才生态细分为三大梯队：“构建者”（Builders）、“协调者”（Orchestrators）与“赋能专业人员”（Enhanced Professionals）。其中，“构建者”涵盖核心 AI 研究员、机器学习工程师和数据科学家，该群体处于技术核心，但人才池体量最小、招聘竞争最为白热化，且候选人评估决策周期漫长。相比之下，由 AI 产品经理、技术项目经理、自动化专家及数据分析师构成的“协调者”，是连接模型能力与商业价值的关键纽带，也是目前增长最快、企业实现业务落地的最佳招聘切入点。最后，“赋能专业人员”包含了软件工程师、解决方案架构师乃至营销人员，他们正利用 GitHub Copilot 或 Claude 等工具辅助日常工作。Lorenz 提醒，这一群体是企业内部最大且未被充分利用的人才储备，却常常因为官方职位头衔缺少特定 AI 关键词而遭到招聘人员的忽视。地理迁移趋势与非科技行业需求 (5:49) 借助 Lightcast 的劳动力市场情报数据，Lorenz 分析了全美 AI 职位的地域分布与行业流动。尽管加利福尼亚州在招聘职位的绝对总量上继续保持领先，但其他区域的百分比增速更为显著。其中，得克萨斯州的人工智能职位同比增长率已经超越加州，而明尼苏达州则呈现出 110% 的同比增长势头。在行业分布上，招聘需求正从传统的硅谷科技巨头向实体和重工业领域迅速扩散。制造业、医疗保健和航空航天等传统产业正在积极招揽 AI 专业技术人员以推进业务现代化，并与科技大厂展开直接竞争。与此同时，德勤（Deloitte）、埃森哲（Accenture）和普华永道（PwC）等专业咨询机构培养了大量具备企业级项目经验的工程师，这些人才对能够提供直接业务所有权和可衡量影响力的企业内部岗位展现出浓厚兴趣。Dice 调查进一步证实，71% 的 AI 专业人员对传统或非大型科技公司持开放态度，为寻求自动化与预测性维护的实体产业创造了引才契机。算法筛选悖论与招聘流程重构 (8:07) Lorenz 警告称，企业在招聘环节对自动化筛选工具的过度依赖，正在严重削弱其雇主品牌吸引力。Dice 的调查数据显示，68% 的技术专业人员对完全由 AI 主导的招聘流程表示不信任，46% 的候选人明确表示如果提供替代方案将选择退出自动化简历筛选。最理解 AI 技术的求职者反而对招聘场景下的 AI 算法最具怀疑态度。目前，82% 的技术求职者使用生成式 AI 优化简历与求职信，且 51% 并不担心被检测，导致招聘流程异化为“求职者 AI 与企业筛选机器人相互博弈”的低效循环。招聘环节的另一误区是死守机械的职位描述而忽略候选人的学习潜力。部分雇主开出要求“具备 15 年 ChatGPT 经验”的不切实际条件，不仅破坏了企业信誉，也因僵化的关键词过滤错失了适应型人才。例如，几年前曾被视为独立专长的“提示词工程”如今已快速演化为基础技能。此外，资深 AI 人才需要时间评估团队文化与技术架构，单方面追求 30 天极速招聘往往适得其反。为破解这一难题，部分招聘人员采用定制化提问并借助 AI 过滤套话，以此识别出真正具备专业理解力和业务敏捷度的候选人。应对高流动性与构建留存优势 (13:41) AI 专业人才对自己所掌握的市场筹码具有清晰认知，这导致该群体呈现出显著的高流动性。在 Dice 的调研中，60% 的 AI 专业技术人员对自身获得理想职位充满信心，而整体技术群体的这一比例仅为 36%；同时，约 75% 的 AI 从业人员计划在接下来的 12 个月内跳槽，明显高于一般技术员工 65% 的跳槽意向。Lorenz 提示企业密切关注核心技术人员的离职前兆，例如退出内部技术研讨、对新项目产生抵触情绪或在日常团队协作中逐渐边缘化。在吸引与留存策略上，高水平人才看重的是薪酬区间的透明度、清晰的面试流程、实际的技术自主权，以及打造能够投入生产环境的系统而非临时性原型。为了确立竞争优势，企业管理层应当通过内部圆桌会议等形式，直接向现有技术团队探询其留职意愿的核心动因，明确自身在业务使命与技术责任感上的独特性，同时提供充足的学习预算以帮助员工适应工具链的快速演进。 Lorenz 总结强调，AI 战略的核心起点不是算法模型而是人才队伍。面对激烈的外部竞争与繁重的申请量，企业既不能完全依靠人工审核，也不能全盘转向冰冷的自动化筛选，而必须采纳兼顾规模效率与关键触点温度的“混合招聘”模式。由于具体技术工具可以通过在职培训掌握，但文化契合与协作信任难以短期弥补，组织应以透明度和真实人际互动为基石，通过尊重、挑战与技术归属感来维系团队的长期凝聚力。

Detailed notes — zh

## Background 该内容由 Ai4 于 2025 年 8 月 24 日发布在 YouTube 平台。页面简介显示，本次演讲由 DHI Group 旗下科技招聘平台 Dice 的市场研究与洞察总监 Brooke Lorenz 主讲，旨在基于技术人才调研与招聘趋势大数据，为企业高管提供在激烈竞争中构建高绩效 AI 团队、吸引并留存顶尖人才的实战指南与避坑策略。 ## TL;DR 演讲指出企业 AI 转型真正的核心瓶颈是人才而非算力与模型，近半数投入 AI 的企业内部极度缺乏落地与运维专业人员。Brooke 将 AI 人才生态划分为构建者、编排者与赋能型专业人士三类，并指出过度依赖冰冷的自动化筛选和刻板死板的职位描述是导致人才流失的致命误区。面对 AI 时代双向博弈的招聘现状，企业必须依靠薪酬与流程的高透明度建立信任，运用人机协同的混合模式筛选深度思考者，并通过赋予技术自主权、展现业务影响力和发挥传统行业的稳定性优势来吸引并留住顶尖人才。 ## Atmosphere & style 演讲以轻松幽默的自嘲式台风在拉斯维加斯现场开场，随后迅速切入扎实硬核的调研数据与行业洞察，整体呈现出务实、睿智且极具说服力的专业氛围。演讲者语言风趣犀利，在剖析企业死板 JD 和“AI 招 AI”荒诞博弈时一针见血，既有对宏观人才版图迁移的全局审视，又充满对技术人真实心理诉求的人文关怀。 ## Key takeaways - [01:26](https://www.youtube.com/watch?v=Qhr1zALLZao&t=86s) 企业在 AI 转型中普遍重资金与算力投入，但贝恩研究表明近半数投资 AI 的企业内部缺乏落地人才，AI 战略的核心瓶颈是人而非技术本身。 - [01:26](https://www.youtube.com/watch?v=Qhr1zALLZao&t=86s) 全美近 40% 的科技招聘需求已包含 AI 技能，模型的能力上限最终取决于背后的人才水准。 - [03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) 破除高价从大厂挖角的刻板印象，AI 人才生态由构建者（Builders）、编排者（Orchestrators）和赋能型专业人士（Enhanced Professionals）三层梯队构成。 - [03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) 编排者（如 AI 产品经理、数据分析师）是当前增长最快且性价比极高的发力点；赋能型专业人士则是内部 Upskilling 最理想的庞大人才池。 - [05:49](https://www.youtube.com/watch?v=Qhr1zALLZao&t=349s) AI 人才地域与行业版图正在迁移：德州和明尼苏达州招聘增速迅猛，制造、医疗、航空航天及管理咨询公司正与硅谷巨头展开正面争夺。 - [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) 过度针对职位描述（JD）死抠头衔或强推 30 天速成冲刺是常见招聘误区，企业应更看重候选人的思维流畅度与敏捷适应潜力。 - [11:23](https://www.youtube.com/watch?v=Qhr1zALLZao&t=683s) Dice 调研显示 68% 的技术专业人士不信任完全由 AI 驱动的招聘流程，纯自动化筛选容易导致优秀人才反感并主动退出。 - [11:39](https://www.youtube.com/watch?v=Qhr1zALLZao&t=699s) 求职端有 82% 的技术人才使用 AI 工具辅助申请，招聘方需通过定制化深度问题并配合大模型聚类，筛选出具备独立思考的差异化回答。 - [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) 薪资透明化与合理的任职要求是吸引顶尖人才的基础，企业必须向候选人公开招聘流程中 AI 介入的具体环节与规则以建立信任。 - [15:01](https://www.youtube.com/watch?v=Qhr1zALLZao&t=901s) AI 专业人士拥有极高流动性：近 60% 确信能迅速找到新工作，近 75% 计划在一年内跳槽，企业需警惕技术骨干淡出交流等早期脱节信号。 - [15:01](https://www.youtube.com/watch?v=Qhr1zALLZao&t=901s) 顶尖 AI 人才更看重战略级业务影响力、技术自主掌控权、持续学习预算以及健康的组织文化，而非单纯为实验室原型抛光打磨。 - [17:59](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1079s) 71% 的 AI 人才对非科技大厂持开放态度，传统制造业可凭借职业稳定性、实体场景落地机会和直接业务归属感构建差异化竞争优势。 - [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) 最理想的招聘是人机协同的“混合模式”：利用 AI 提高处理效率，同时在关键评估与决策节点保留不可替代的人际沟通与温度。 - [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) AI 战略始于人而非模型，企业的招聘与待人方式决定了团队构成，进而决定了 AI 业务落地的最终上限。 ## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=Qhr1zALLZao&t=4s) 开场：AI 狂热背后的“重资产、轻人才”断层演讲者 Brooke Lorenz（来自 DHI Group 旗下科技招聘与职业平台 Dice 的市场研究与洞察总监）以幽默的方式开场。她调侃作为一个“内向的研究人员”，在拉斯维加斯的讲台上演讲略带挑战，并且笑称 Dice 可不是赌桌上掷的骰子，而是一个专注于连接技术人才与雇主、提供职业建议和求职机会的专业平台。她深知自己可能是阻碍大家前往会后 Happy Hour 的“绊脚石”，因此直入主题，向台下观众抛出一个在拉斯维加斯极具共鸣的场景：在过去这一年里，几乎所有企业的高管会议中都出现过“我们需要在 AI 上加大投资”的声音。然而，企业目前的普遍做法是将海量预算砸向数据、算法模型和基础设施等底层技术栈，却常常忽略了最核心的要素——真正让这些系统运转起来的“人”。Brooke 指出，这种对硬件与算力过度聚焦、对人才规划相对脱节的现象，正是当下企业 AI 转型中最致命的盲区。 ### [01:26](https://www.youtube.com/watch?v=Qhr1zALLZao&t=86s) AI 转型的真正瓶颈是人才而非技术跨行业企业正在为 AI 系统投入数以百万计的资金，但贝恩咨询（Bain & Company）的一项研究显示，近半数投资 AI 的公司公开承认，自己内部根本缺乏落地和运维这些系统所需的专业人才。与此相对的是，AI 在极短时间内从前沿小众领域迅速跃升为行业主流需求，目前全美近 40% 的科技职位招聘需求中都已明确包含 AI 技能。这一剧烈分化揭示了当前矛盾的核心：AI 应用落地的加速推进导致市场对人才的需求爆发式增长，而具备实战经验的资深从业者供给却远远跟不上。企业最终陷入了“买得起先进系统，却招不到人运转”的尴尬境地。因此，AI 转型的真正瓶颈并非算法和算力，而是人才。企业的核心课题不应仅仅停留在“如何部署更多 AI”，而必须转向“如何吸引、评估并留住那些能够将 AI 愿景转化为现实的人才”。正如 Brooke 强调的，"你的模型再强大，其上限也完全取决于模型背后的人"；而这群掌握核心生产力的人才不仅拥有海量选择，而且正在密切审视企业是否真正懂得如何做优秀的 AI 研发。 ```mermaid flowchart LR accTitle: AI Talent Bottleneck Flow accDescription: 从资金投入到人才瓶颈的业务逻辑链路 subgraph Input ["技术投入"] A["巨额技术栈投资 算力/模型/数据"] end subgraph Gap ["现实断层"] B["40% 科技岗位要求 AI 技能"] C["近 50% 企业缺乏内部落地人才"] end subgraph Reality ["核心瓶颈"] D["真正瓶颈：人才而非算力"] E["核心挑战：吸引/评估/留用人才"] end A --> B A --> C B & C --> D D --> E ``` ### [03:38](https://www.youtube.com/watch?v=Qhr1zALLZao&t=218s) 破除“高价挖人”迷思：AI 人才生态的三层架构许多企业在规划 AI 招聘时容易陷入单一视角的刻板印象，总联想到 Meta 开出数百万乃至上亿美元天价从 OpenAI 挖角顶尖研究员的新闻。然而，现实中的 AI 人才需求远比这丰富得多。Brooke 将整个 AI 人才生态系统清晰地拆解为三个不同梯队，指出各类角色分工明确、各司其职： ```mermaid flowchart TD accTitle: AI Talent Ecosystem accDescription: AI人才生态的三种角色划分及其特征 subgraph Ecosystem ["AI 人才生态系统架构"] B["构建者 Builders AI 研究员、ML 工程师、数据科学家 • 资源极度稀缺、竞争白热化 • 招聘决策周期长、审视全面"] O["编排者 Orchestrators AI 产品经理、项目经理、自动化专家、数据分析师 • 连接技术与业务需求 • 当前增长最快、性价比最佳的甜蜜点"] E["赋能型专业人士 Enhanced Professionals 现有软件开发者、架构师、使用 Copilot/Claude 的营销人员 • 规模最大但常被忽略的潜在池 • 适合内部技能重塑（Upskilling）"] end B --> O O --> E ``` - **构建者（Builders）**：包括核心 AI 研究员、机器学习（ML）工程师和具备深厚技术底蕴的数据科学家。这是市场上存量最小、争夺最激烈的群体。他们拥有漫长的招聘考察周期，面对雪片般的邀约，他们会严谨细致地评估企业的技术底座、使命愿景与团队契合度。 - **编排者（Orchestrators）**：包括 AI 产品经理、AI 项目经理、自动化专家以及数据分析师。他们负责架起商业目标与技术能力之间的桥梁，既懂客户真实痛点，又理解模型的能力边界。这一群体是目前招聘市场上增速最快的细分赛道，也是企业性价比极高的发力点。 - **赋能型专业人士（Enhanced Professionals）**：由熟练使用 GitHub Copilot、Claude 等 AI 工具的现有软件工程师、解决方案架构师乃至市场营销人员构成。这是市场上体量最大却最容易被企业忽视的群体，仅仅因为他们的职位名称里没有“AI”字样。他们是企业推行内部技能重塑（Upskilling）和内部人才流动的最理想来源。 ### [05:49](https://www.youtube.com/watch?v=Qhr1zALLZao&t=349s) 寻找 AI 人才的新版图：地域迁移与跨界竞争基于劳动力市场智能分析工具 Lightcast 对全美科技职位招聘大数据的追踪，Brooke 揭示了 AI 人才供需版图正在发生的显著变化，为非科技巨头企业指出了差异化揽才的破局点：在地域分布上，加利福尼亚州在 AI 招聘总量上依然无可争议地拔得头筹；但在同比增速方面，德克萨斯州已经反超加州，明尼苏达州的 AI 岗位招聘量同比增速更是达到了惊人的 110%。在行业分布上，争夺 AI 人才的早已不仅限于 Meta 和 Google 等硅谷科技巨头。制造业、医疗健康（Healthcare）以及航空航天（Aerospace）等传统与高端制造行业对 AI 人才的需求正大幅攀升，开始与传统科技大厂正面争抢同类技术人员。此外，德勤（Deloitte）、埃森哲（Accenture）、普华永道（PwC）等管理咨询公司持续保持着极高的 AI 人才招聘强度，沉淀了一批具备跨行业、企业级大型 AI 系统落地经验的高级专家。对于非科技巨头企业而言，只要愿意放低身段、切合候选人实际诉求并展示出清晰直接的业务成效与特色，当前正是从这些渠道吸引人才的最佳窗口期。 - **德州与明尼苏达州**：德州 AI 招聘同比增速超越加州，明尼苏达州同比增幅达 110%。 - **非科技大厂崛起**：制造、医疗、航空航天正在成为技术人才的主要竞聘方。 - **咨询行业人才池**：咨询顾问积累了多行业的复杂落地经验，常在寻求能够产生直接、可衡量业务结果的新机会。 ### [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) 常见招聘误区一：过度针对职位描述（JD）而非人的潜力进行优化企业在触达候选人的招聘流程中往往漏洞百出，甚至造成恶劣的第一印象。例如一位资深 ML 工程师应聘一家宣称“AI 优先”的公司，结果简历直接被简历解析 Bot 自动无视，或者在提交申请仅仅 10 分钟后就收到系统自动下发的拒信。这种冰冷生硬的自动化流程不仅暴露出公司 AI 成熟度的低下，更让求职者产生严重的反感。在 Dice 针对技术专业人员的调研中，一位求职者直言不讳地评价道："这看起来根本不是什么前沿科技，而是显得无比愚蠢。这说明你们根本不懂这项技术，那我凭什么要为你工作？" 围绕这一现象，Brooke 深入剖析了企业最常犯的第一个招聘错误——**过度针对职位描述（JD）进行刻舟求剑式的优化，却忽视了人的适应能力与综合潜力**： | 常见招聘痛点 | 产生的根本原因与危害 | 优秀企业的应对解决方案 | |:--- |:--- |:--- | | **死抠特定职位头衔** | 技能迭代极快（例如 Prompt Engineer 几年前是专项职位，如今已变成通用基础能力），容易漏掉高潜力复合型人才。 | **看重思维流畅度与敏捷适应力**，而非单纯看候选人是否掌握某款特定工具。 | | **强行推进 30 天速成招聘冲刺** | 顶级 AI 人才手握多个 Offer，需要充分时间了解企业的 Tech Stack、团队文化与业务使命，强压节奏只会导致候选人流失。 | **拉长评估周期、给予充分沟通空间**，让候选人深度了解团队与技术挑战。 | | **直接复制粘贴通用 JD** | 粗制滥造的 JD 向市场释放出“招聘方完全不懂这个岗位、不知道成功标准为何物”的负面信号。 | **招聘团队与 Hiring Manager 深度绑定**，定制 AI 专属评估维度，将 JD 视为展示企业文化与愿景的“核心宣传案（Pitch）”。 | ### [11:23](https://www.youtube.com/watch?v=Qhr1zALLZao&t=683s) 常见招聘误区二：过度依赖自动化招聘工具许多企业试图用自动化工具来解决招聘效率问题，却引发了候选人强烈的信任危机。Dice 最新的调研数据显示，**68% 的技术专业人士明确表示他们不信任完全由 AI 驱动的招聘流程**，更有 46% 的受访者对纯自动化筛选持强烈的保留态度。这种对自动化工具的过度依赖，正直接将最优质的技术人才推向竞争对手。 ### [11:39](https://www.youtube.com/watch?v=Qhr1zALLZao&t=699s) 招聘中的“AI 对轰”与破局之道当被给予选择权时，多数技术人才宁可主动选择退出 AI 自动化筛选环节。这就催生了招聘领域的一个奇妙悖论：企业正试图用 AI 工具去招聘最懂 AI 的技术人才，而这些人才恰恰对这类筛选工具抱有最根深蒂固的不信任感。最理解 AI 运作机制与局限性的人，反而是对招聘中使用 AI 最持怀疑态度的人。演讲者指出，这并不是呼吁招聘团队彻底弃用 AI，毕竟当下面试官和猎头每天都被海量的简历淹没，确实亟需工具辅助提效，但关键在于“绝不能在招聘中丢失至关重要的人际连接”。事实上，这场“算法军备竞赛”并非企业单方面的行为，求职者同样深谙此道。调查数据显示，高达 82% 的技术专业人士在求职和投递流程中使用了 AI 工具，且有 51% 的人完全不担心招聘官会识别出他们使用了 AI。这就导致招聘流程演变成了一种荒诞的景象——求职者的 AI 代理与企业的 AI 筛选工具在互相“对话”。为了在层层 AI 生成的同质化噪音中筛选出真正有思考的候选人，一位资深招聘主管分享了他的实用策略：向候选人提出 5 个高度定制化的具体问题，随后将所有回复导入 ChatGPT，让模型比对并归类出千篇一律的套路回答与具有独特视角的回答，最终只将那些展现出真正思考深度的差异化候选人推进到下一轮。 ```mermaid flowchart TD accTitle: 招聘流程中的人机混合互动模式 subgraph CandidateSide ["求职端 (82% 使用 AI)"] C1["候选人撰写申请"] --> C2["利用 AI 生成/优化简历与回答"] end subgraph ScreeningProcess ["筛选与博弈层"] C2 --> S1["AI 与 AI 的同质化信息碰撞"] S1 --> S2["提出定制化深度问题"] S2 --> S3["利用 LLM 聚类过滤平庸套话"] end subgraph DecisionSide ["招聘端 (保留人际连接)"] S3 --> D1["筛选出具独立思考的差异化回答"] D1 --> D2["人工介入与深度沟通"] end ``` - **82% 的技术专业人员** 在求职与申请环节中使用 AI 工具。 - **51% 的求职者** 并不担心招聘方能够检测出其使用了 AI。 - 破局核心在于**提出针对性极强的问题**，并利用大语言模型筛选出非标准化、真正动脑思考的独特见解。 --- ### [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) 吸引顶尖人才的基石：透明度与真实需求在明确了筛选环节的痛点后，接下来的核心问题是：究竟是什么能够真正吸引顶尖 AI 人才并让他们愿意留下来？调研反馈中最核心的诉求非常一致——技术人员极度渴望“透明度”。这种透明度首先体现在薪资范围的公开发布上；其次，职位要求必须切合实际。如果在职位描述中荒谬地要求候选人“具备 15 年 ChatGPT 经验”，专业人才会一眼看穿企业的不专业与敷衍。此外，企业必须清晰地向候选人公开招聘全流程，尤其是其中涉及 AI 筛选的环节与规则。 AI 领域的从业者比任何人都清楚自动化工具的软肋与死板机制。正如调研中一位受访者直言：“AI 并非万无一失，它可能会漏掉高素质的候选人，因为 AI 的配置逻辑只懂得机械地寻找特定关键词。”另一位技术人员则补充道：“技术人的职业生涯很少是完全线性的，而这种宝贵的复合型经验和复杂背景，在自动化筛选中往往被直接抹杀。”正因如此，公开透明的沟通机制是建立信任的第一步。 - **薪资透明化** 与 **合理的任职要求** 是吸引顶尖人才的基础底线。 - 拒绝荒谬的职位要求（例如脱离技术发展历史的夸张工作年限要求）。 - 自动化筛选容易错失非线性职业路径的高潜力人才，企业需向候选人公开 AI 介入的具体环节与流程。 --- ### [15:01](https://www.youtube.com/watch?v=Qhr1zALLZao&t=901s) 薪酬之外的留才法则与离职预警信号薪酬对于顶尖人才固然重要，但这仅仅是基础考量的一部分。顶尖 AI 从业者真正看重的，是自己能否在业务中产生**战略级影响力**、能否获得持续成长的空间、是否拥有开展最佳工作的灵活性，以及自己是否在“亲手构建真正落地的产品，而不仅仅是在为实验室原型抛光打磨”。因此，企业在招聘宣讲时应重点阐述自身业务的战略价值故事，并展示健康的团队文化与成长通道。从留存现状来看，AI 人才市场呈现出极高的流动性与跳槽信心。数据显示，近六成（约 60%）的受访 AI 专业人士坚信自己能在市场上迅速找到心仪的新职位，而其他普通技术人员中这一比例仅为 36%；更有近四分之三（约 75%）的 AI 专业人士计划在未来一年内更换雇主（其他技术人员该比例为 65%）。这种极高的市场需求和人才流失风险要求企业必须制定深思熟虑的留才策略。 ```mermaid flowchart LR accTitle: AI 核心人才留存四大支柱 subgraph RetentionPillars ["人才留存四大支柱"] A["明确的业务影响力 (拒绝纯原型打磨)"] B["成长与学习预算 (跟进前沿技术迭代)"] C["自主权与业务挑战 (解决有价值的难题)"] D["健康团队文化 (推崇好奇心/伦理/跨团队影响)"] end ``` 留存工作往往始于招聘阶段的体验，因为招聘方式直接映射出公司的管理与运营风格。如果在面试中对候选人“已读不回”（Ghosting），他们就会预料到入职后自己也会被领导层忽视；如果让候选人经历 10 轮繁琐流程却得不到任何实质反馈，他们就会认定未来的晋升评估也是这般官僚。与此同时，管理者必须警惕团队内部早期的“隐性脱节信号”：如果技术骨干开始退出技术论坛交流、对新项目产生抵触情绪、或是逐渐从团队协作中抽离，这往往意味着他们已经萌生去意。 - **59% 的 AI 专业人员** 对找到更好的工作充满信心（普通技术人员为 36%）。 - **近 75% 的 AI 专业人员** 计划在接下来的一年内跳槽（普通技术人员为 65%）。 - **早期离职预警信号** 包括：淡出技术社群交流、消极对待新项目、回避跨部门协作等。 - **留才关键要素**：清晰呈现工作对业务指标的推动、提供持续的技术学习预算、赋予解决核心难题的自主权、以及推崇包容与好奇心的组织文化。 --- ### [17:59](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1079s) 打破大厂光环：传统行业的差异化竞争优势许多非科技行业的管理者常常担忧自己无法在人才争夺战中与科技大厂（Big Tech）竞争，但调研数据却带来了利好消息：仅有 29% 的 AI 专业人士表示他们只倾向于去科技大厂工作，这意味着有高达 71% 的人才对其他类型的企业和行业持完全开放的态度。诸如制造业等传统行业虽然往往不在 AI 人才求职的第一选择清单上，但实际上在自动化、预测性维护和数字化转型等场景中对 AI 人才有着极为迫切的需求，并且坐拥未被发掘的竞争优势。传统企业破局的核心在于找准并放大自身的独特差异点。一位制造业人才总监曾坦言：“我们最大的阻碍是外界的刻板印象。大家觉得制造业无法与大厂竞争。”但随后她指出了传统企业能提供的一项核心杀手锏——那就是“稳定性”。无论身处哪一行业，企业都应当明确自身的竞争优势，将技术岗位的职责与公司的宏大使命相绑定，强调技术人员在业务落地中的全权自主掌控力（Technical Ownership）。如果不确定自身的潜在优势在哪里，最有效的方式就是直接与现有的技术团队开展圆桌座谈，深入了解究竟是什么让他们愿意长久留下来。 - **71% 的 AI 人才** 愿意考虑非科技大厂的其他类型企业，科技巨头并非唯一选择。 - 传统工业与制造业在自动化及预测性维护方面具备庞大的落地场景。 - 传统企业在与大厂竞争时，可重点强调**职业稳定性**、**清晰的业务落地归属权**以及**直接赋能实体产业的使命感**。 --- ### [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) 核心总结：AI 战略始于人而非模型演讲在尾声部分给出了最具指导意义的论断：“AI 战略从来不是从模型开始的，它是从‘人’开始的（AI strategy doesn't start with models. It starts with people）。”企业招聘与对待人才的方式，决定了团队的构成，而团队的构成最终决定了业务所能实现的一切。演讲者将全场分享提炼为四项核心建议： 1. **顶尖 AI 人才拥有充足的选择权且深知自身价值**：若想赢得他们的青睐，企业必须拿出独具吸引力、能带来实质影响力的发展机会。 2. **在竞争激烈的市场中，信任与透明度远胜过盲目追求速度**：尽管招募人才迫在眉睫，但在招聘流程中花时间与候选人建立深度、真实的信任关系至关重要。 3. **最理想的招聘模式是人机协同的“混合模式”（Hybrid）**：面对浩瀚的申请量，纯人工筛选难以为继；但完全交由 AI 筛选又会摧毁信任并错失良才。应当让人工智能处理其擅长的数据聚类与流程加速，同时在关键节点保留无可替代的人际温情与深度对话。 4. **文化认同与技术硬实力同等重要**：工具与技术栈可以后天培训，但文化契合度却难以强求。优秀的人才只会留在那些让他们感到被重视、受尊重并能持续迎接挑战的环境中。 ## Notable quotes > [00:04](https://www.youtube.com/watch?v=Qhr1zALLZao&t=4s) 我们需要在 AI 上加大投资 — Brooke Lorenz > [01:26](https://www.youtube.com/watch?v=Qhr1zALLZao&t=86s) 你的模型再强大，其上限也完全取决于模型背后的人 — Brooke Lorenz > [08:07](https://www.youtube.com/watch?v=Qhr1zALLZao&t=487s) 这看起来根本不是什么前沿科技，而是显得无比愚蠢。这说明你们根本不懂这项技术，那我凭什么要为你工作？ — 求职者（调研受访者） > [11:23](https://www.youtube.com/watch?v=Qhr1zALLZao&t=683s) 68% 的技术专业人士明确表示他们不信任完全由 AI 驱动的招聘流程 — Brooke Lorenz > [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) AI 并非万无一失，它可能会漏掉高素质的候选人，因为 AI 的配置逻辑只懂得机械地寻找特定关键词 — 技术专业人士（调研受访者） > [13:41](https://www.youtube.com/watch?v=Qhr1zALLZao&t=821s) 技术人的职业生涯很少是完全线性的，而这种宝贵的复合型经验和复杂背景，在自动化筛选中往往被直接抹杀 — 技术专业人士（调研受访者） > [15:01](https://www.youtube.com/watch?v=Qhr1zALLZao&t=901s) 亲手构建真正落地的产品，而不仅仅是在为实验室原型抛光打磨 — Brooke Lorenz > [17:59](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1079s) 我们最大的阻碍是外界的刻板印象。大家觉得制造业无法与大厂竞争 — 制造业人才总监 > [19:33](https://www.youtube.com/watch?v=Qhr1zALLZao&t=1173s) AI 战略从来不是从模型开始的，它是从‘人’开始的 — Brooke Lorenz ## Mentioned Brooke Lorenz, DHI Group, Inc, Dice, Ai4, 贝恩咨询 (Bain & Company), Meta, OpenAI, Google, GitHub Copilot, Claude, Lightcast, 德勤 (Deloitte), 埃森哲 (Accenture), 普华永道 (PwC), ChatGPT ## Video description Brooke Lorenz, Director, Market Research and Insights, DHI Group, Inc. Abstracts For C-suite leaders beginning their AI journey, this session offers research-backed insights into building high-performing AI teams in a fiercely competitive talent market. Drawing on new survey data from tech professionals and hiring trends from Dice, you’ll learn how to attract, evaluate, and retain top AI talent by building trust, while avoiding the strategic missteps that cause organizations to lose them. [Source](https://www.youtube.com/watch?v=Qhr1zALLZao)

Source quotes

[0:04] Brooke Lorenz: 我们需要在 AI 上加大投资 [1:26] Brooke Lorenz: 你的模型再强大，其上限也完全取决于模型背后的人 [8:07] 求职者（调研受访者）: 这看起来根本不是什么前沿科技，而是显得无比愚蠢。这说明你们根本不懂这项技术，那我凭什么要为你工作？ [11:23] Brooke Lorenz: 68% 的技术专业人士明确表示他们不信任完全由 AI 驱动的招聘流程 [13:41] 技术专业人士（调研受访者）: AI 并非万无一失，它可能会漏掉高素质的候选人，因为 AI 的配置逻辑只懂得机械地寻找特定关键词 [13:41] 技术专业人士（调研受访者）: 技术人的职业生涯很少是完全线性的，而这种宝贵的复合型经验和复杂背景，在自动化筛选中往往被直接抹杀 [15:01] Brooke Lorenz: 亲手构建真正落地的产品，而不仅仅是在为实验室原型抛光打磨 [17:59] 制造业人才总监: 我们最大的阻碍是外界的刻板印象。大家觉得制造业无法与大厂竞争 [19:33] Brooke Lorenz: AI 战略从来不是从模型开始的，它是从‘人’开始的

Source quotes

[8:07] Survey Respondent: It doesn't look cutting edge. It looks foolish. It shows you don't understand the technology. Why would I want to work for you? [13:41] Survey Respondent: AI is configured to look for specific things [13:41] Survey Respondent: careers in tech are rarely linear and that complexity gets lost in automated screening. [19:33] Brooke Lorenz: AI strategy doesn't start with models. It starts with people.

Original transcript

[0:04] Hi everyone. Well, this isn't [0:06] intimidating at all for an introverted [0:08] researcher. [0:09] Thank you all so much for joining my [0:11] session. Um, I understand I'm probably [0:14] standing in your way of happy hour, so I [0:16] appreciate your time here and spending [0:18] the next 20 minutes with me. So, as [0:21] Kamorei introduced, my name is Brooke [0:23] Laoren, and I work for a company called [0:25] DICE, which is not the company that [0:28] produces what you're thrown out there on [0:30] the craps table. Um, but instead, we are [0:32] a tech career marketplace. And we [0:35] connect employers with skilled tech [0:37] professionals and we also provide tech [0:39] professionals with a career advice and [0:42] job opportunities as they navigate [0:44] throughout their tech career. [0:46] Now I want to start this session by [0:49] sharing a scenario that might sound [0:51] familiar to all of you. [0:54] I am willing to bet seeing as how we are [0:56] in Vegas that at some point this year [0:58] you have been in a meeting or a [1:00] conversation where somebody said we need [1:03] to invest more in AI. [1:06] And chances are the conversation focused [1:08] more so on the data and the models and [1:10] the infrastructure but perhaps not the [1:13] the talent that's actually needed to [1:15] bring all of those systems to life. And [1:17] that disconnect is what I will share [1:19] with all of you today and talk about [1:26] sorry um across industries companies are [1:30] investing millions in tech stacks needed [1:33] for AI systems. But here's the thing. [1:36] Even though they are investing millions, [1:39] according to a company or excuse me, [1:41] according to a study by Bay & Company, [1:44] nearly half of those same companies [1:46] admitted that they don't have the [1:48] in-house expertise required to bring [1:50] those systems to life. At the same time, [1:53] AI has gone from being niche to [1:56] mainstream almost overnight. [1:59] And we're now seeing AI skills be a [2:02] requirement in nearly 40% of tech job [2:05] postings in the US. So this dynamic [2:08] captures the heart of the issue. AI [2:11] adoption has accelerated so rapidly that [2:14] demand for talent has really taken off. [2:17] But the supply of the experienced [2:19] practitioners needed hasn't quite kept [2:22] up. So the result is these organizations [2:25] are investing in systems that they're [2:27] not yet staffed to support. [2:30] And that means the real bottleneck in AI [2:33] transformation isn't just technology, [2:36] it's talent. [2:40] And so the question isn't so much how do [2:42] we deploy more AI, but how do we [2:45] attract, evaluate, and retain the people [2:47] who can deliver on our AI vision? Your [2:51] models are only as powerful as the [2:54] humans behind them. And those humans, [2:57] they have options. They are in high [2:59] demand. They are mobile and they are [3:02] paying attention to how you are hiring [3:04] and how you are communicating with them [3:07] and whether you actually understand what [3:09] it takes to do great AI work. And so [3:12] today we're going to dig into what it [3:15] takes to build high performing AI teams [3:18] in a fiercely competitive market. We'll [3:21] look at what an AI talent ecosystem [3:24] looks like. [3:26] I'll share a couple of common hiring [3:28] mistakes for AI talent and how to design [3:31] an AI talent strategy that builds trust [3:34] and aligns with the future that you are [3:36] building. [3:38] Because when most companies think of AI [3:40] hire, they're picturing one type of [3:42] person, right? We've all heard the story [3:45] of Meta's poaching offer of $und00 [3:47] million for OpenAI's top researcher, but [3:51] the reality is it's much more nuanced [3:53] than that. So let me break down what the [3:56] talent ecosystem actually looks like [3:59] because not all AI roles are create [4:01] created equal. [4:05] So first we have our builders and these [4:08] are your core AI researchers the very [4:10] person that I was just talking about in [4:12] my meta example. Then you have ML [4:14] engineers and data scientists who have [4:17] deep technical skills. This is the [4:20] smallest pool and also they are in [4:24] they're in such high demand that there's [4:25] a lot of competition to try to get these [4:28] professionals. They also have a long [4:30] hiring cycle, right? They have because [4:33] they're in such high demand, they are [4:35] getting inundated with all of these [4:36] different offers, but they're taking the [4:38] time to really go through each and every [4:40] offer to make sure it's the right fit [4:42] for them. [4:44] And next, you have orchestrators. [4:47] These are roles like AI product and [4:50] program managers, automation specialists [4:52] and data analysts. These professionals [4:55] bridge business and technical goals and [4:58] they also understand both customer needs [5:00] and model capabilities. Um we are seeing [5:03] this be the fastest growing segment and [5:06] it's a a pretty good sweet spot right [5:08] now for hiring in the AI realm. [5:12] And finally enhanced professionals. This [5:15] is your existing software developers and [5:17] engineers, solution architects, and even [5:20] marketers who are using tools like [5:22] C-pilot and Claude. [5:25] This is actually the largest untapped [5:27] pool and also serves um they're ideal [5:31] for upskilling in internal mobility, [5:34] but they often get overlooked simply [5:36] because they don't have AI in their [5:38] title. So, if you're only hiring for the [5:41] builders, you may be missing the talent [5:44] that brings AI to life in real world [5:46] settings. [5:49] Okay. So, now I've talked about who you [5:51] need to hire. The next question becomes, [5:53] where can you find this talent and where [5:56] are some opportunities that other [5:57] companies might be missing? Before I [6:00] share the data on the next couple of [6:01] slides, I just want to make a quick [6:02] comment. The analysis that I will share [6:05] was done by looking at US tech job [6:07] postings using a labor market [6:09] intelligence tool called Litecast. And [6:12] so this is really giving you a snapshot [6:14] of where the demand is um and also where [6:17] companies are actively investing. [6:20] So we'll start with location. [6:23] Now it'll be a surprise to exactly [6:24] nobody in here when I tell you [6:26] California is leading the way when it [6:28] comes to AI job posting volume. But what [6:32] might surprise you is knowing that Texas [6:35] actually outpaces California when we [6:38] look at year-over-year growth in AI job [6:41] postings. [6:42] Minnesota is also another AI hiring [6:45] growth that saw 110% year-over-year [6:48] growth. [6:51] Another shift that we are seeing is that [6:53] it's not just Meta and Google hiring for [6:56] AI professionals. Industries like [6:59] manufacturing, healthcare, and aerospace [7:02] are now significantly increasing their [7:04] demand for AI talent and competing for [7:08] tech roles or excuse me, competing with [7:10] tech firms for these very roles. [7:13] So, if you're not in big tech, this is [7:16] your window of opportunity. [7:19] um and that you can especially attract [7:21] them if you're willing to meet them [7:22] where they are and highlight what makes [7:25] your organization stand out which I'll [7:27] share more about in just a few slides. [7:30] Um another sector that is shaping AI [7:32] hiring right now is consulting. So firms [7:35] like Deote, Accenture, PWC, they are all [7:39] consistent consistently maintaining [7:42] um high AI hiring. [7:45] Um and so they're creating the steady [7:47] pipeline of professionals [7:49] um and professionals in this industry [7:51] have helped build enterprisecale AI [7:53] systems often in a wide range of [7:55] industries. Um, and these individuals [7:58] might be looking for their next [7:59] opportunity, especially if you can offer [8:01] them um direct meaningful outcomes. [8:07] Okay. Um, I want to shift focus now and [8:10] talk about hiring because even if you [8:12] know exactly who you need, there's still [8:14] a question of how you're reaching them. [8:16] And this is where some organizations may [8:18] be falling apart. So, I want you to [8:21] imagine for a second that you are a [8:23] machine learning engineer looking for [8:25] your next role. You apply to a company [8:28] claiming to be AI first only to be [8:31] ghosted by a resume bot or you're left [8:34] scratching your head knowing you have [8:36] all the qualifications, but 10 minutes [8:38] after you submitted your application, [8:40] you automatically got a rejection. And [8:43] this this experience alone could shed [8:46] light on a company's AI maturity and is [8:48] a disconnect that candidates are [8:50] sensing. [8:52] In a survey that I recently conducted [8:54] where I asked tech professionals, "What [8:57] is your perception of a company that [8:59] uses AI in their hiring process? [9:02] This is what one candidate told us. It [9:05] doesn't look cutting edge. It looks [9:07] foolish. It shows you don't understand [9:09] the technology. Why would I want to work [9:11] for you? And something like that hurts. [9:15] It could hurt if you're a recruiter [9:16] sitting in here. I understand that. But [9:19] the reality is this this comment was not [9:22] an outlier. We heard it in a few [9:24] instances. [9:25] And they're not wrong. So let me show [9:27] you where AI hiring might fall apart. [9:31] Mistake number one, optimizing for job [9:34] descriptions, not human potential. [9:39] The problems listed out here. Um, if [9:42] you're only searching for titles, then [9:44] you could miss an evolving skill set. [9:46] So, if you think about prompt engineers, [9:48] for example, that was a specialized role [9:51] only a few years ago, but it has since [9:54] evolved into being more of a baseline [9:56] skill that we were seeing in job in job [9:58] postings. [10:00] Second here is rushing 30-day hiring [10:03] sprints fail. I mentioned earlier top AI [10:07] talent is in such high demand and [10:09] they're getting multiple opportunities. [10:12] So they they need time to assess your [10:14] tech stack, your mission, and the team [10:16] that they will be working on. They can't [10:18] just do it all in 30 days. [10:21] And third, copy and pasting job [10:23] descriptions signal that you don't quite [10:26] understand the role and what it what [10:28] success could look like. So what are [10:31] some things companies are doing to fix [10:33] these problems? Number one, they're [10:35] hiring for fluency and adaptability over [10:38] tool mastery. Number two, they partner [10:41] very closely with hiring managers. Some [10:44] of the most successful recruiting [10:45] stories I hear is when recruiters and [10:48] hiring managers have a tight [10:49] relationship and they get each other. [10:52] The recruiter works to really understand [10:54] what the needs and wants are for a [10:55] particular role so they know how to be [10:58] successful. [11:00] Third is creating AI specific evaluation [11:03] criteria. And last, treat your job [11:06] descriptions [11:08] um in a way that showcases your culture [11:10] and intent. Don't just treat them like a [11:13] legal disclaimer because in your whole [11:15] hiring process, your whole hiring [11:18] process is ultimately your pitch. [11:23] Mistake number two, overreiance on [11:26] automated tools. Get this. In our [11:29] survey, 68% [11:31] of tech professionals said they distrust [11:34] fully AIdriven hiring. 46% would [11:39] actually opt out of AI screening if they [11:41] were given the choice. [11:43] So, we're using AI tools to hire AI [11:46] talent who fundamentally distrust those [11:49] tools. The people who understand AI best [11:53] are the ones who are most skeptical of [11:55] it in hiring. Now, I'm not saying don't [11:59] use AI at all in the hiring process. I [12:02] recognize recruiters are inundated with [12:04] the amount of applications they receive [12:07] and they need these tools to help them. [12:09] But what I am saying is don't lose that [12:11] human connection that is so critical in [12:14] hiring. [12:16] I also recognize this story has two [12:18] sides. [12:20] There are some recruiters who are guilty [12:22] of using AI, sure, but the candidates, [12:25] they're just as guilty. And you can see [12:27] that by the numbers on this slide which [12:29] shows 82% of tech professionals in our [12:32] survey said they use AI in job in the [12:36] job search and application process. And [12:38] the top use cases are listed there to [12:40] the right. And 51% they're not all that [12:44] concerned that a recruiter might detect [12:47] their use of AI. [12:49] So then if recruiters are using AI and [12:51] candidates are using AI, then it just [12:54] becomes this conversation between AI [12:56] tools, right? And then what happens? How [12:59] can a recruiter really cut through all [13:01] of the AI noise? [13:04] I asked this question recently to a [13:06] recruiter and he gave me this advice. He [13:08] said, I like to ask my candidates five [13:12] very specific and tailored questions and [13:14] I included his favorite up here on the [13:16] slide. He then takes all the responses [13:19] he gets, puts it into chat GPT, and asks [13:22] it to identify the responses that are [13:24] very similar in nature and the responses [13:27] that are very different. And then he [13:29] advances the ones that are different [13:30] enough that shows they're actually [13:32] paying attention. Um, advances them to [13:34] the next round in the hiring process. So [13:38] take that for what you will. [13:41] Okay. So I've talked about the hiring [13:43] process. So now the next question [13:45] becomes what actually draws top AI [13:47] talent in and what makes them stay? [13:51] And I'll start with what we hear most [13:53] from tech professionals in that they [13:55] want transparency. This means publishing [13:58] your salary ranges, sharing realistic [14:01] job requirements. Right? If you're [14:02] asking somebody if they have 15 years of [14:05] chat GPT experience, they're going to [14:08] see right through that and know that's [14:09] not you're not paying attention. And [14:11] lastly, absolutely outline your hiring [14:14] process, especially if it involves using [14:17] AI [14:19] because the people that you're hiring [14:21] for these AI roles, they understand [14:23] these tools better than anyone and they [14:25] know their limitations. As you can see [14:27] from this example quote from our same [14:30] survey where this person said, "AI is [14:33] not foolproof and can miss quality [14:36] candidates because AI is configured to [14:38] look for specific things." [14:42] Here's another example quote where this [14:44] person goes on to say, "Care careers in [14:47] tech are rarely linear and that [14:49] complexity gets lost in automated [14:52] screening." [14:54] This is why transparency matters. [15:01] Okay, now let's talk about what makes [15:03] your opportunity stick in a candidate's [15:06] mind. Yes, salary matters. I'm not going [15:08] to stand up here and tell you it [15:09] doesn't. It absolutely does. But for top [15:12] AI talent, it's only part of the [15:14] equation. What they really want to know [15:16] is will they have strategic impact? Will [15:20] they grow? Do they have the flexibility [15:23] to do their best work? and are they [15:25] building something real and not just [15:27] polishing a prototype. So start with [15:29] your strategic value story and also [15:32] focus on culture and their growth [15:34] opportunities. [15:38] Now, I realize attracting AI talent [15:41] could be difficult and retaining them [15:43] can be even more difficult, especially [15:46] when you consider the stats on this [15:48] slide that show nearly six and 10 of the [15:51] AI professionals in our survey said they [15:53] are confident they could find a [15:55] favorable role versus just 36% of other [15:59] tech professionals. and nearly [16:01] threearters told us they plan to change [16:03] employers in the next year versus 65% of [16:08] other tech professionals. So this kind [16:11] of high demand and high risk or highf [16:14] flight risk might seem like a recipe for [16:16] disaster, but if you can really focus on [16:18] your retention strategy and be [16:20] thoughtful about it, then you might just [16:23] keep your top tech talent around for [16:25] longer. [16:28] And that should start with the way that [16:30] you hire since that can translate into [16:32] how you lead and run your business. So [16:34] if you ghost candidates during hiring, [16:37] they might expect to be overlooked [16:39] later. If you make them jump through 10 [16:41] steps with no feedback, they might [16:44] expect that when it comes time for a [16:45] promotion as well. And watch for these [16:48] early signs of disengagement. If you [16:51] start noticing your tech team is [16:53] disengaging with technical forums or if [16:56] they're pushing back on new projects or [16:58] withdrawing from collaboration, these [17:01] might all be signs that someone on your [17:03] tech team has a foot out the door [17:04] already. [17:06] So, what keeps them clear business [17:08] impact? If you can really show them how [17:11] their work is moving the needle, this [17:13] could help keep them around for longer. [17:16] Career growth and learning budgets is a [17:18] huge one. I don't have to tell anybody [17:20] in here. Tech evolves nearly every day [17:22] and changes constantly. And if your tech [17:26] team is not keeping up with that or [17:28] you're not allowing them to keep up with [17:29] that, that could send them out the door. [17:33] Um, third is flexibility and challenge. [17:37] Give them the autonomy that they need [17:39] and meaningful problems to solve. And [17:42] last is culture. [17:44] really ensuring that culture is one that [17:48] values curiosity and ethics and [17:51] influence across teams. This could all [17:54] be what builds long-term commitment with [17:56] your tech team. [17:59] Now, I know some leaders, they might [18:01] worry that they can't compete with big [18:03] tech, but hopefully I offer up some good [18:06] news here. In our survey, we had 29% of [18:09] AI professionals tell us they prefer to [18:11] work for a big tech company. But what [18:13] that also means is 71% are open to other [18:17] business types. In fact, some of them [18:20] some of the most overlooked industries [18:22] like manufacturing for example are [18:25] sitting on a competitive advantage, but [18:27] they might not realize it since they're [18:29] often not top of mind for AI talent. But [18:32] industries like manufacturing, they [18:34] absolutely need AI talent for automation [18:38] and predictive maintenance and digital [18:40] transformation. [18:42] But the thing these companies may not be [18:44] taking into consideration is really [18:46] leaning into their competitive [18:48] advantage. For example, when I spoke to [18:50] a director of talent at a manufacturing [18:52] company last year, she said, "Our [18:55] biggest struggle is perception. Nobody [18:57] thinks of manufacturing because we [18:58] cannot compete with the big tech [19:00] companies." But then she highlighted a [19:02] differentiator that her company could [19:05] offer, which was pretty big at the time [19:06] was a year ago. And she said, "We can [19:08] offer stability." So, no matter what [19:11] industry you're in, get clear on your [19:14] edge. Make the connection to mission. [19:17] Um, emphasize your technical ownership [19:20] and find your hidden strengths. If you [19:22] don't know what your hidden strengths [19:23] are, talk to your tech team and do a [19:27] round table with them simply to [19:28] understand from them what what keeps [19:30] them around. [19:33] So, what does this all mean in practice? [19:35] If you take anything away from this [19:38] presentation, let it be this. [19:40] AI strategy doesn't start with models. [19:43] It starts with people. [19:46] And the way you hire shapes the team [19:49] that shapes everything else. And here's [19:51] a few other top takeaways. [19:54] Top AI talent, they have options. I've [19:56] mentioned that already. They know their [19:58] worth. So if you really want to win them [20:00] over, your opportunity has to stand out. [20:03] Number two, trust in transparency beats [20:06] speed in competitive markets. I know you [20:09] want to move fast to hire that AI [20:11] talent, but really take time to [20:13] cultivate those relationships with them [20:15] during the hiring process. [20:18] Number three, the best hiring approach [20:21] for both candidates and companies is [20:23] hybrid. It's not all human. It can't be [20:26] because that's not sustainable the [20:28] amount of applications they are getting. [20:30] But it also can't just be AI for the [20:32] reasons I mentioned earlier. It has to [20:35] be both. Use the technology for what [20:37] it's good at, but keep that human [20:40] connection alive. That is so important [20:42] in the hiring process. [20:45] And finally, cultural alignment matters [20:47] just as much as technical skill, [20:49] especially for retention. You can teach [20:52] tools, but you can't necessarily teach [20:54] culture fit. So, keep in mind, people [20:57] want to stay where they feel seen and [20:59] challenged and respected. [21:03] With that, I want to thank you all for [21:05] your time. [21:07] Thank you.