In conversations about AI, capital spending and platform selection often receive more attention than human capability. Yet data from the World Economic Forum’s Future of Jobs Report 2025 suggests that skills gaps are a primary barrier to business transformation. This is not merely an HR issue. It is becoming a condition for an AI strategy to move from experimentation into operations.
Published in January 2025, the report draws on responses from more than 1,000 employers representing over 14 million workers across 55 economies. Surveyed employers expect 39% of the workforce’s core skills to change by 2030; 63% identify skills gaps as a key barrier to transformation; and 77% expect to upskill or reskill their workforce. These figures do not predict what will happen in every individual company, but they indicate an important shift: AI is no longer only a technology-team story.
From buying tools to designing operating capability
An organization can purchase access to an AI tool quickly. The capability to turn that tool into dependable results develops much more slowly. Employees need to know how to provide context, evaluate output, identify missing information, protect data and place AI results inside an accountable workflow. Managers need to select suitable problems, set quality standards and decide when automation is inappropriate. When both capabilities are absent, usage may be high while value remains fragmented.
Skills development should therefore begin with real work rather than a generic class on features. A sales team may need to practice preparing opportunity reports and checking customer data. An operations team may need to learn how to handle exceptions in an AI-supported process. Middle managers may need the capability to read metrics, assess risk and coordinate responsibility between people and systems. The same tool requires different capabilities across roles.
Training matters only when it is connected to work redesign
Completing a course is not evidence of transformation. The more useful questions are: which work is done differently after training, which standard improves, and which result is monitored? The WEF report also highlights the simultaneous importance of technology skills such as AI, data and cybersecurity with analytical thinking, creativity, resilience, leadership and collaboration. This reminds employers that an “AI skills” list cannot be separated from management work and cross-functional coordination.
Implication for business
Instead of setting the target of “AI training for every employee,” an organization can select three priority processes, identify the roles involved and define a specific capability for each role. It can then measure correct process adoption, processing time, errors and user feedback. This approach is smaller than a communication campaign, but it allows leaders to see where skills investment is creating value.
Three management decisions need to be made early
First, identify which capabilities are shared and which must be function-specific. Asking good questions, checking sources and protecting data are useful across many groups. But applying AI to demand forecasting, customer service or risk analysis requires different domain knowledge. Second, assign owners for capability development rather than leaving training entirely to one unit. Third, link learning to work choices and implementation metrics instead of evaluating it by attendance.
The labour market will continue to change, but companies do not need to wait for a fully certain picture before acting. A sensible starting point is to treat skills gaps as an operating risk and design a learning rhythm connected to workflows. In AI strategy, human capability is not work that follows technology. It is the condition that allows technology to produce a management result.
References
World Economic Forum. (2025). The Future of Jobs Report 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/
