Workers in developing economies face significantly lower risks of job displacement from generative artificial intelligence than their counterparts in high-income countries, but persistent infrastructure and skills gaps could limit the productivity gains available from the technology, the World Bank has said.
The Group’s World Development Report 2026: The Promise of Artificial Intelligence found that 4.5 percent of existing jobs in low and middle-income countries are at risk of automation by generative AI, compared with 14.2 percent in high-income economies.
Despite the lower exposure to automation, the potential for AI to improve productivity in developing economies is relatively close to that of wealthier countries.
The report estimated that 16.2 percent of jobs in developing economies could experience meaningful productivity gains from AI, compared with 18.7 percent in high-income countries.
The World Bank said this indicates that the most significant opportunity for developing economies may lie in using AI to strengthen the productivity of existing workers rather than replacing them.
AI could help doctors, farmers, businesses and public institutions overcome some of the capacity constraints that have limited service delivery in developing countries.
The technology could support healthcare workers in diagnosis, assist farmers with agricultural decisions and enable businesses to improve productivity. Governments could also use AI to strengthen areas such as tax collection, social programmes, disaster response, healthcare and education.
However, the World Bank warned that the ability to convert these potential gains into economic growth will depend heavily on whether developing countries can address basic infrastructure and capacity deficits.
Africa’s infrastructure challenge
The challenge is particularly significant in sub-Saharan Africa, where access to electricity and reliable internet remains limited in many communities.
The World Bank said nearly one-third of rural schools in the region still lack reliable electricity, while more than two-thirds lack dependable internet access.
The deficits could restrict access to AI tools and limit the ability of schools, businesses, communities and public institutions to participate in the emerging digital economy.
The Bank said efforts to expand electricity access would therefore remain central to the region’s ability to benefit from AI.
It pointed to Mission 300, an initiative being implemented with partners to provide energy access to 300 million people across sub-Saharan Africa by 2030.
According to the World Bank, closing the electricity and connectivity gaps would provide a foundation for broader digital and AI inclusion.
Developing economies also need greater access to computing capacity, relevant data and technical skills to translate AI’s potential into productivity gains.
Growth opportunity at risk
The AI challenge comes as developing economies contend with their weakest average growth performance in three decades.
The World Bank said AI could help improve economic performance before the end of the 2020s while delivering tangible benefits to citizens.
However, it warned that countries could fail to capture these gains if they do not address the conditions required for effective adoption.
Advanced AI development remains concentrated among a relatively small number of countries and companies, while many developing economies lack the infrastructure, data, skills and institutions needed to use the technology effectively.
The Bank warned that this could result in AI widening existing gaps between countries rather than helping poorer economies close them.
It said unchecked AI adoption could also increase inequality within countries and concentrate market power, while creating risks for public trust, safety, individual rights and social cohesion.
Gaurav Nayyar, director of the World Development Report 2026, said developing countries have a limited window to establish the foundations needed to benefit from AI.
“The window to get this right is narrow,” Nayyar said, describing AI as an opportunity to tackle problems that have remained unresolved for generations.
He said countries that invest in electricity, connectivity, skills and institutions would be better positioned to ensure that AI delivers benefits to their populations.
