World Bank Says Developing Countries Could Gain More From AI Than Richer Economies

Artificial intelligence is often presented as a threat to workers in developing countries. But a major new World Bank assessment reaches a more nuanced conclusion: developing economies may have more to gain from AI than advanced economies, provided they build the infrastructure, skills and institutions needed to use it effectively.

The World Bank's World Development Report 2026, focused on artificial intelligence and development, argues that AI could help developing countries accelerate economic transformation by improving productivity, expanding access to expertise and helping businesses overcome shortages of skilled workers.

The finding comes with an important caveat. AI could also deepen inequality between countries if computing infrastructure, skills and access to advanced models remain concentrated in a small number of richer economies.

Developing countries may face less job disruption

According to the World Bank analysis cited by African Business, less than 10% of jobs in developing economies are susceptible to AI automation, compared with more than one-third in high-income economies. At the same time, about 16.2% of jobs in low- and middle-income countries could be complemented by AI, meaning the technology improves workers' capabilities rather than replacing them.

In many poorer economies, employment is concentrated in occupations involving substantial physical, interpersonal or manual work — areas today's AI systems are generally less capable of automating completely. The bigger immediate opportunity may therefore be helping existing workers become more productive.

AI could help developing countries catch up

Training a doctor, engineer or teacher can take years or decades. AI systems, by contrast, can make certain forms of expertise available almost instantly. An AI system could help a small business owner analyse financial information, help a teacher adapt educational material, or give workers access to specialised information that previously required an expert.

Small businesses could be among the biggest winners

In many developing economies, a large share of employment is generated by very small firms and self-employed workers. The World Bank says these businesses could start benefiting from AI without building sophisticated AI infrastructure themselves, because AI is increasingly incorporated into tools they already use — messaging applications, business software and social media.

• Marketing and content production
• Customer service and first response
• Accounting and administration
• Sales follow-up and research

“Small AI” could be particularly important

Not every developing country needs to build enormous frontier-AI systems. The World Bank increasingly highlights Small AI — more affordable, accessible AI applications designed around specific problems and capable of running on everyday devices such as smartphones.

Examples include agricultural advice, translation, education, business administration, customer service, health-support tools, financial services and government services. The opportunity is not to build the world's largest model. It is to apply existing AI effectively to local problems.

Kenya offers an interesting example

Kenya is specifically mentioned as an example of AI-assisted public-sector transformation. The World Bank has worked with Kenya's judiciary on algorithmic systems designed to improve the assignment of cases to mediators based on previous performance. Sometimes AI's value comes simply from making an existing administrative process faster and more data-driven.

But there is a major risk: the AI divide

The most advanced models, chips and data centres are concentrated in a relatively small number of companies and countries. A country may have access to an AI model but lack the computing infrastructure to customise it. Businesses may depend entirely on foreign platforms, and local languages and cultural contexts may be poorly represented.

Africa is home to more than 2,000 languages, but many leading AI models have comparatively limited training data for African languages. That creates an opportunity for African developers: local AI systems adapted for African languages, industries and social contexts could provide value generic global models cannot.

Infrastructure could become the biggest bottleneck

The World Bank describes the foundations through its 4Cs framework: Connectivity (reliable digital networks), Compute (access to processing power), Context (relevant data, applications and algorithms) and Competency (the skills required to develop and use AI).

Energy is now part of AI policy too. Data centres consume large amounts of electricity and can require substantial water for cooling, which creates a difficult calculation for countries with constrained infrastructure.

The risk to developing-country jobs has not disappeared

The report warns that developing economies could lose some traditional advantages in call centres, entry-level software development, finance and business-process outsourcing. That makes skills development increasingly important — in AI-assisted software development, data analysis, digital marketing, workflow automation, cybersecurity, AI governance and industry-specific AI applications.

The World Bank says AI-related jobs are already growing faster in middle-income countries than in high-income countries, while generative AI skills are spreading beyond traditional ICT occupations.

What this means for Kenyan businesses

Many Kenyan businesses are small and still rely on manual administrative processes. AI does not have to replace a large workforce to have a meaningful economic impact. A small business could use AI to respond to customers automatically, generate marketing content, prepare quotations, analyse sales, manage appointments, summarise documents, automate follow-ups and support employees with research.

Kenya's realistic opportunity is in application, not just infrastructure: agriculture, financial services, healthcare, education, retail, government services and professional services all have local problems AI can address today.

Our take

AI is neither automatically a disaster for developing countries nor a guaranteed shortcut to prosperity. A country with strong connectivity, electricity, skills and institutions can use AI to increase productivity. A country without those foundations may simply become dependent on foreign AI systems.

The World Bank's central message is best understood as an opportunity with a deadline: developing countries may have a lot to gain from AI, but they need to build the foundations now to capture those gains.

Sources: World Bank World Development Report 2026, Digital Progress and Trends Report 2025: Strengthening AI Foundations, World Bank research on AI readiness and development, and reporting by African Business.