In 2025, LinkedIn reported that the most in-demand skill globally was “AI literacy” — not a technical specialization, but the baseline ability to understand and use AI tools effectively. That same year, the World Economic Forum estimated that 44% of workers’ core skills would be disrupted by AI and automation by 2027. And the Bureau of Labor Statistics projected that occupations requiring AI skills will grow 13% faster than the overall labor market through 2032.
The message is clear: AI skills are not a niche requirement for data scientists. They are a workforce-wide imperative. And the United States, despite leading in AI development, has a significant AI skills gap that threatens organizational competitiveness, economic productivity, and individual careers.
À retenir
- 44% of workers' core skills will be disrupted by AI and automation by 2027 (World Economic Forum)
- Only 30% of US workers have received formal AI training, despite 75% of businesses planning AI adoption
- The skills gap is not just technical — AI literacy, critical thinking, and data fluency are the most needed skills
- Companies like JPMorgan, Amazon, and Google are investing billions in workforce AI reskilling programs
The data: How wide is the gap?
The numbers paint a stark picture:
Bureau of Labor Statistics (2025): Occupations in AI-related fields are projected to grow at 13% annually through 2032, compared to 3% for the overall labor market. But the pipeline of qualified workers is not keeping pace.
Pew Research Center (2025): 70% of US workers say they have not received any formal training on AI tools, despite the majority using them (often as shadow AI).
World Economic Forum Future of Jobs Report (2025): 44% of worker skills will be disrupted by 2027. The report identifies AI and big data as the number-one skill set driving job transformation.
Amazon/Accenture (2025): 72% of US employers report difficulty finding workers with AI skills, up from 49% in 2023.
McKinsey Global Institute (2025): Up to 30% of hours currently worked in the US could be automated by generative AI by 2030 — affecting 12 million workers who may need to transition to different occupations.
44%
of workers' core skills will be disrupted by AI and automation by 2027
Source : World Economic Forum Future of Jobs Report, 2025
Which skills are most needed?
The AI skills gap is not just about learning to code or build machine learning models. It spans five categories:
1. AI literacy (everyone)
The foundation. Every employee needs to understand:
- What AI can and cannot do
- How to evaluate AI outputs critically
- When to trust and when to question AI
- Basic prompt engineering for generative AI tools
- Data privacy and AI governance basics
This is not technical training. It is the modern equivalent of computer literacy in the 1990s — a baseline competency for the entire workforce.
2. Data fluency (most roles)
AI runs on data. Workers need:
- Understanding of data quality and its impact on AI outputs
- Ability to interpret data-driven insights
- Knowledge of data privacy and handling requirements
- Statistical reasoning sufficient to evaluate AI claims
- Awareness of bias in data and models
3. Critical thinking and verification (most roles)
As AI generates more content, analysis, and recommendations, the ability to evaluate AI outputs becomes critical:
- Hallucination detection — recognizing when AI invents facts, citations, or data
- Source verification — independently confirming AI-provided references
- Bias recognition — identifying when AI outputs reflect training data biases
- Contextual judgment — knowing when AI recommendations fit the specific situation
4. Domain-specific AI application (role-specific)
Different functions need different AI skills:
- Marketing: AI content generation, personalization, analytics
- Finance: AI-powered forecasting, fraud detection, compliance monitoring
- Healthcare: Clinical AI, ambient documentation, patient data AI
- Legal: Contract analysis, research, risk assessment
- HR: AI-assisted recruiting, compliance with anti-discrimination law
- Operations: Process automation, predictive maintenance, supply chain optimization
5. AI strategy and governance (leadership)
Executives and managers need:
- Understanding of AI’s strategic implications for their industry
- Ability to evaluate AI investments and ROI
- Knowledge of AI governance frameworks and regulatory requirements like the NIST AI RMF
- Change management skills for AI transformation
- AI policy development and enforcement capability
The biggest mistake organizations make is treating the AI skills gap as a technical problem requiring technical training. The largest gap is in AI literacy and critical thinking — skills that apply to every role and require a fundamentally different training approach than engineering upskilling.
What leading companies are doing
JPMorgan Chase
JPMorgan is investing $2 billion annually in technology training, with AI as a central focus. The bank’s approach:
- Mandatory AI literacy training for all 300,000+ employees
- An internal AI platform (LLM Suite) that gives employees hands-on experience with approved tools
- A dedicated AI research team that develops training curricula based on real internal use cases
- Partnerships with universities for advanced AI education
Amazon
Amazon committed $1.2 billion to upskilling 300,000 workers through its “Upskilling 2025” program, with AI and machine learning as a major component:
- Machine Learning University, originally internal, now open to external learners
- AI Ready initiative providing free AI skills training to 2 million people globally by 2025
- Internal reskilling programs that have moved warehouse workers into technical roles
Google is investing $75 million in AI training through its Google.org foundation:
- Google AI Essentials certificate program for general workforce AI literacy
- Integration of AI skills into existing Google Career Certificates
- Partnerships with community colleges and workforce development organizations
- Free AI training tools and curricula for employers
AT&T
AT&T’s $1 billion workforce reskilling initiative is a case study in proactive transformation:
- Identified that 100,000 employees (nearly half the workforce) were in roles that would be significantly changed by AI and automation
- Built an internal reskilling platform combining online learning, mentorship, and project-based experience
- Result: 70% of affected employees have successfully transitioned to new roles
$2B
invested annually by JPMorgan Chase in technology training, with AI as the central focus
Source : JPMorgan Chase Annual Report, 2025
Why reskilling beats hiring
Many organizations try to close the AI skills gap by hiring. The math does not work:
Supply is limited. There are not enough AI specialists to go around. The demand-supply gap for AI talent in the US is estimated at 500,000+ positions (Korn Ferry, 2025).
Costs are prohibitive. AI specialists command salaries 30-50% above market average for comparable roles. For most organizations, hiring enough specialists to cover every function is not financially viable.
Culture matters. External hires bring technical skills but lack institutional knowledge. Internal reskilling builds AI capability within employees who already understand the business, the customers, and the organizational context.
Speed is critical. Hiring cycles take months. A well-designed AI training program can upskill existing employees in weeks.
The most effective approach is both/and: hire a small number of AI specialists to lead, and reskill the broader workforce to follow.
Start your reskilling program with “AI champions” — employees who are already experimenting with AI tools (often as shadow AI). They are your most motivated learners and your best internal advocates. Train them first, then empower them to train their teams.
The role of regulation
US regulators are beginning to connect the AI skills gap to compliance:
- NIST AI RMF: The Govern function explicitly calls for workforce AI competency as a governance requirement.
- Colorado AI Act (2026): Requires organizations deploying high-risk AI to have staff with appropriate training and competency.
- FTC guidance: Organizational liability for AI misuse implies a duty to train employees.
- SEC expectations: AI risk disclosures should address workforce readiness — a skills gap is a material risk.
- Industry regulators: Banking (SR 11-7), healthcare (FDA guidance), and other sector regulators increasingly expect documented AI competency.
Close the AI skills gap with Brain
Brain is purpose-built to close the AI skills gap at scale. Practical, role-specific AI training covering literacy, tool competency, risk awareness, governance compliance, and advanced application. Modules designed for how adults actually learn — short, scenario-based, immediately applicable. A dashboard that tracks competency, not just completion.
Whether you are training 50 employees or 50,000, Brain gets your workforce ready. Explore our plans to get started.
Related articles
AI in the Workplace: What Is Actually Changing (2026)
Understand how AI reshapes jobs, creates new roles, and widens the skills gap. UK workforce data and practical guidance for managing the transition.
AI Change Management: 5-Phase Adoption Playbook
Overcome resistance, build communication strategies, phase training and measure AI adoption. Practical change management playbook.
AI Decision Making: 5 Ways Leaders Make Better Choices
Use AI for scenario planning, risk assessment, and bias mitigation. A practical guide to human-AI collaboration in executive decision making.