World Bank shifts focus to AI-resilient industries to boost employment

The World Bank is rethinking its approach to job creation in the world's poorest regions, taking into account how artificial intelligence is transforming the labor market. The organization wants to more strongly support industries where technology complements people rather than quickly displacing them.

Why Strategy Is Changing

Until recently, development programs could be built around simple logic: investment flows in, business expands, employment grows. Now this chain no longer looks so straightforward. AI is increasingly taking on tasks in office-based, service, and analytical professions, which means international development institutions must anticipate which jobs will retain demand in a few years and which might face rapid automation.

For the World Bank, this is not a theoretical debate but a practical question for countries where job creation remains the main condition for poverty reduction. If previously the emphasis could be on simply scaling any labor-intensive sector, now it's more important to look at employment sustainability. Otherwise, a program might attract funding and even show quick initial results, but then run into the fact that some functions can be performed more cheaply by software or AI services.

Which Industries Win

AI-resilient sectors are usually understood as those requiring physical presence, local context, manual labor, trust, and work in unpredictable environments. This does not mean rejecting technology. Rather, the point is to invest in areas where AI increases human productivity instead of eliminating the need for people altogether.

For the poorest regions, this approach is especially important, because the labor market there is sensitive to any sharp shifts.

• construction and local infrastructure
• agriculture and processing
• basic healthcare and care
• repair, maintenance, and logistics
• education and applied training

These sectors share a common feature: they are tied to the real local economy and often require a combination of skills that are difficult to fully automate. Even if AI takes on planning, diagnosis, documentation, or staff training, the final work still falls to people. So the bet shifts not away from AI, but toward models where technology acts as an employment accelerator rather than a direct replacement.

AI as a New Filter

If this logic takes hold in World Bank programs, it will change not just the list of priority sectors but also how projects are evaluated in practice. The question will be: how many jobs will an initiative create today, and how sustainable will those jobs be in a world where AI becomes cheaper and more accessible?

This could affect the structure of loans, grants, retraining programs, and project selection criteria for private capital. For developing economies, there is also a broader signal here. Competing solely on cheap labor becomes riskier if some global companies can automate back-office functions, customer service, quality control, or parts of digital production. This means employment policy must be built not around an abstract number of vacancies, but around how much these jobs are embedded in local infrastructure, supply chains, real people's needs, and local demand.

What This Means

The World Bank is effectively acknowledging that the era when almost any employment program could be automatically considered useful is ending. Now the key question is not just how many jobs are created, but how compatible they are with AI and protected from rapid automation. For the poorest countries, this could mean a shift from quantitative employment growth toward more precise, sustainable, and long-term industrial policy.