Talent and Skills Transformation in the AI Era

What talent and skills does the AI era demand? The answer affects an organization’s competitiveness in AI transformation. McKinsey’s 2024 survey lists talent shortages, alongside inaccuracy, among the risks organizations encounter when pursuing value from generative AI.[1] This section explains the skills needed in the AI era, a strategic framework for acquiring talent, and reskilling case studies from major companies.

Skills Needed in the AI Era

Skills needed in the AI era are frequently misunderstood as only “AI technical skills,” but in reality they are composed of three layers: technical skills, AI literacy, and human skills.

graph TD
    subgraph "AI Era Skills Pyramid"
    H["Human Skills\nCritical thinking, ethical judgment, creativity,\ncommunication, leadership"]
    A["AI Literacy\nBasic understanding of AI mechanics,\nprompt design, AI output evaluation,\nAI risk awareness"]
    T["Technical Skills\nML/DL, data science,\nMLOps, AI architecture,\nprogramming"]
    end
    H --- A --- T

Skill Layer	Target Audience	Core Capabilities
Technical Skills	Data scientists, AI engineers, ML engineers (~5–10% of all employees)	Model development, MLOps, data infrastructure design
AI Literacy	All employees (regardless of role or level)	Using AI tools, evaluating AI output, recognizing risks
Human Skills	All employees (increasingly critical in the AI era)	Critical thinking, ethical judgment, creativity, interpersonal relationships

Gartner (2024) predicts that “AI literacy will become a foundational capability for all business professionals, just like digital literacy.”[3]

What Is AI Literacy and Why Does Every Employee Need It?

AI Literacy is the ability to understand the basics of how AI works, use AI tools appropriately, critically evaluate AI output, and recognize AI-related risks. It does not include programming or machine learning expertise.

Components of AI Literacy

Component	Content	Example
Conceptual understanding	Basics of what AI is and how generative AI works	Understanding that LLMs generate output probabilistically
Practical ability	Effectively using AI tools in daily work	Prompt design, using AI output, iterative improvement
Critical evaluation	Evaluating the accuracy, bias, and limitations of AI output	Identifying hallucinations (AI generating incorrect information)
Risk awareness	Recognizing and handling AI-related risks	Basic understanding of confidential information input risks and copyright issues

Why Every Employee Needs It

There are three reasons AI literacy is needed by all employees:

1. AI tools are permeating all job functions — AI integration in business tools like Microsoft Copilot, Google Workspace AI, and Salesforce AI is advancing rapidly
2. Humans bear responsibility for judging AI output quality — Documents, analyses, and proposals generated by AI still require final human evaluation; those without evaluation skills risk being misled by AI
3. Organizations without AI literacy carry higher risk — Risks including inappropriate input of confidential information, uncritical adoption of AI output, and copyright infringement increase

Build / Buy / Borrow / Bot (4B) Framework

The 4B Framework (Build / Buy / Borrow / Bot) is a decision-making framework for strategically combining four options for acquiring AI talent and capabilities. A concept promoted by Accenture and Gartner, it has become the standard thinking tool for talent strategy in AI transformation.[2][3]

graph TD
    Q["How to secure the capabilities needed?"] --> B1["Build\nInternal development (reskilling)"]
    Q --> B2["Buy\nExternal hiring"]
    Q --> B3["Borrow\nOutsourcing and consulting"]
    Q --> B4["Bot\nReplace with AI and automation"]
    B1 --> C1["Takes time but\nbecomes embedded in culture"]
    B2 --> C2["Immediately effective but\nhigh cost and competition"]
    B3 --> C3["Speed and flexibility,\nknowledge transfer is a challenge"]
    B4 --> C4["Cost-efficient,\nchallenges with high-precision tasks"]

Build: Internal Development (Reskilling)

Build is an internal development approach that has existing employees acquire AI-related skills.

Item	Content
Best suited for	Existing employees with domain knowledge and organizational context; mid-career and senior employees
Main methods	Systematic training programs, OJT, mentoring, hackathons
Advantages	Cultural fit, combination with domain knowledge, long-term cost efficiency
Disadvantages	Takes time (6–18 months); not all employees can make the transition
Key prerequisite	Learning opportunities, time, and psychological safety must be secured

Buy: External Hiring

Buy is an approach of hiring AI specialists from outside.

Item	Content
Best suited for	Roles requiring high technical skills (AI engineers, data scientists, AI Product Managers)
Advantages	Immediately effective, injects external knowledge and culture, high technical expertise
Disadvantages	Intensifying competition for talent, high cost, takes time to adapt to organizational culture
Market situation	Supply-demand gap for AI talent remains severe as of 2026 (LinkedIn, 2024)[4]

Borrow: Outsourcing and Consulting

Borrow is an approach to temporarily procuring capabilities through consulting firms, startups, freelancers, or outsourcing.

Item	Content
Best for	Filling gaps until internal capabilities are built, specialist participation in specific projects
Advantages	Speed, flexibility, adjustable cost
Disadvantages	Difficult to retain know-how in-house, risk of confidential information management, dependency risk
Key management point	Design knowledge transfer mechanisms and an internalization roadmap in advance

Bot: Replace with AI and Automation

Bot is an approach to replacing business tasks that humans used to handle with AI or automation. This is not synonymous with “headcount reduction” — it is framed as business redesign to concentrate humans on higher-value work.

Item	Content
Best suited for	Business tasks that are highly repetitive, rule-based, or require high-volume processing
Examples	Generating standard documents, data aggregation, initial screening, basic customer service
Key point	Design by distinguishing between automatable “tasks” and “roles” that humans should take

How to Combine the 4Bs

The 4Bs are not mutually exclusive options — they are a portfolio strategy to be used in combination.

graph LR
    subgraph "Phase 1 (Early AI Transformation)"
    P1["Borrow-heavy\nConsultant and partner utilization"]
    end
    subgraph "Phase 2 (Scaling Phase)"
    P2["Buy + Build in parallel\nHiring and development simultaneously"]
    end
    subgraph "Phase 3 (Maturity Phase)"
    P3["Build + Bot focus\nInternal development + expanding automation"]
    end
    P1 --> P2 --> P3

Reskilling and Upskilling in Practice

Reskilling is acquiring entirely new skills to handle a role whose content has changed significantly. Upskilling is developing existing skills to perform one’s current role at a higher level.

Amazon’s Case Study: $700M Investment Program

In 2019, Amazon announced it would invest $700 million to reskill more than 100,000 employees by 2025. This program is a proactive investment anticipating that roles in warehousing, delivery, and call centers will change due to automation and AI.[6]

Key programs:

Program Name	Target	Content
Amazon Technical Academy	Non-technical employees	Support for transition to software engineering
Machine Learning University	Technical employees	Improvement of ML and AI skills
AWS Training & Certification	All employees	Acquisition of cloud and AI basics
Career Choice	Warehouse and logistics employees	Tuition support for college or vocational school while employed

The important lesson from Amazon’s case is “invest proactively before demand rises.” Recognizing that roles will change due to automation and AI, they designed the time and opportunity for affected employees to transition to new careers in advance.

Google’s Case Study: AI Essentials and Company-Wide Rollout

Google offers Google AI Essentials, a public online course that requires no technical experience and teaches foundational skills for using generative AI at work.[7]

Google’s distinctive approach is a thorough commitment to “Learn by Doing.” Programs are centered not on classroom learning but on assignments to apply AI to participants’ actual work.

Microsoft’s Case Study: Copilot Adoption and Integrated Learning

Since the company-wide rollout of Copilot in 2023, Microsoft has implemented the “AI Skills Initiative” that integrates tool provision with learning support.

• Scenario-based Copilot learning content prepared by job function
• Training for managers on “managing teams using AI tools”
• A mechanism that analyzes Copilot usage data and connects low-adoption employees and departments to support

Microsoft and LinkedIn’s 2024 Work Trend Index reports rapid workplace AI adoption and emphasizes the need for organizational guidance and training.[8]

New Roles Emerging in the AI Era

AI transformation not only changes existing roles but creates entirely new ones that didn’t previously exist.

Role	Primary Responsibilities	Required Background
AI Product Owner	Product owner of AI systems, defining and realizing business value	Product management + AI literacy
Prompt Engineer	Input design for LLMs, quality improvement, prompt library management	Language skills + LLM understanding
AI Ethics Officer	Managing ethical risks of AI use, governance design	Ethics and regulatory law + AI literacy
AI Change Manager	Designing and driving organizational transformation accompanying AI adoption	Change management + AI literacy
AI Trainer / RLHF Specialist	Human feedback to AI models, quality improvement	Domain knowledge + AI quality evaluation
MLOps Engineer	Production operations, monitoring, and continuous improvement of AI models	Infrastructure + ML + DevOps
AI Business Analyst	Identifying AI adoption opportunities, ROI estimation, business requirements definition	Business analysis + AI literacy

Many of these roles are “expanded” versions of existing roles born when AI literacy is added. Rather than hiring completely new specialists from outside, it is more realistic for existing product managers, business analysts, and engineers to take on these roles by acquiring AI literacy.

Roles That Will Shrink and How to Transition

AI transformation also means demand will shrink for some roles. Recognizing this and planning the transition support for affected employees is both ethical and practical talent strategy.

Shrinking Work Type	Factors for Replacement	Transition Direction
Creating and editing standard documents	Automated generation by generative AI	Evaluating content quality, overseeing AI output
Data aggregation and report creation	Automation via BI and AI	Data interpretation, involvement in decision-making
Basic customer service	Chatbots and AI dialogue systems	Complex problem solving, emotional support
Standard coding (CRUD, etc.)	AI code generation	Architecture design, code review
Initial screening tasks	AI classification and filtering	Final judgment, exception case handling

The WEF Future of Jobs Report 2025 estimates that macrotrends will create 170 million jobs and displace 92 million jobs by 2030, for a net increase of 78 million.[5] These figures combine technology, environmental, economic, and demographic trends; they are not estimates of AI’s impact alone.

Talent Strategy Pitfalls

Common failure patterns in AI era talent strategy:

Pitfall 1: Over-reliance on Technical Hiring

The most common failure is the misconception that “AI transformation = hiring AI technical talent.” Even if data scientists and ML engineers are hired, AI won’t take root in operations if the business side has low AI literacy. Securing bridge talent (AI Product Owners, AI Business Analysts) and raising company-wide AI literacy are equally or more important.

Pitfall 2: Overconfidence That “Everyone Can Transition with Training”

Reskilling is an effective strategy, but not all employees can acquire all skills. There are individual differences in motivation to learn, aptitude, and environment. Career transition support and retirement support for employees who struggle to transition must also be included in the talent strategy, alongside development efforts.

Pitfall 3: Focusing Only on Short-Term Skill Acquisition

An approach that treats 1–2 days of training as having “learned” AI is often ineffective. Skill practice and retention requires a continuous cycle of learning, applying to actual work, and feedback. One-time training is not sufficient — continuous learning programs must be designed.

Pitfall 4: Hiring AI Talent Without an Environment That Can Use Them

Even if AI specialist talent is hired, they won’t stay in an organizational environment (data infrastructure, decision-making authority, collaborative culture) where they can thrive. Talent strategy must be implemented in conjunction with organizational design and process design.

Summary

Key points in talent and skills transformation in the AI era:

• Skills needed in the AI era are composed of three layers — “technical skills, AI literacy, and human skills” — with AI literacy needed by all employees
• Use the 4B framework (Build/Buy/Borrow/Bot) to design a talent acquisition portfolio tailored to the situation
• Proactive investment in reskilling (like Amazon’s $700M case) is important to implement before role changes occur
• AI transformation creates new roles (AI Product Owner, Prompt Engineer, etc.), many of which are “AI-expanded” versions of existing roles
• Overemphasis on technical hiring, overconfidence in short-term training, and unprepared adoption environments are typical talent strategy pitfalls

FAQ

Q: Where should I start to improve AI literacy? A: Starting with a short program for all employees on “AI literacy basics (what AI is, hallucination risks, basics of prompt input)” is recommended. Using publicly available programs like Google AI Essentials (free) or Microsoft AI Skills (free) can also be effective.

Q: How should I approach reskilling for non-technical departments? A: Rather than learning abstract AI concepts, a practical approach centered on “which parts of my work can I use AI for” is effective. Preparing department-specific case collections and use case libraries that concretize “scenarios changed by AI,” and creating opportunities to actually try tools, leads to lasting adoption.

Q: What should I prioritize when hiring AI talent? A: Beyond technical skills (ML, data science), prioritizing “can they define business problems as AI-applicable questions” and “do they have communication skills with non-technical organizational members” is recommended. AI talent with high technical skills but weak business connections tends to have limited contribution to organizational transformation.

References

1. McKinsey & Company, The State of AI in Early 2024 (2024)
2. Accenture, New Accenture Research Finds that Companies with AI-Led Processes Outperform Peers (2024)
3. Gartner, AI Is Creating New Roles and Skills in Data & Analytics (2024)
4. LinkedIn Economic Graph, The Future of Work Report: AI at Work (2023)
5. World Economic Forum, The Future of Jobs Report 2025 (2025)
6. Amazon, Amazon Pledges to Upskill 100,000 U.S. Employees for In-Demand Jobs by 2025 (2019)
7. Google, Google AI Essentials (2024)
8. Microsoft and LinkedIn, 2024 Annual Work Trend Index (2024)