Small‑scale AI tools poised to boost productivity in emerging economies
In a conversation with CNN’s Fareed Zakaria at Columbia University, World Bank President Ajay Banga emphasized the role of "small AI" in advancing agricultural output in developing nations. Unlike the massive, data‑hungry models dominating the discourse in the United States and Europe, these lightweight predictive solutions operate on modest smartphones and require far less infrastructure.
The Uttar Pradesh case study shows how a voice‑guided app can help a farmer without formal education diagnose a leaf disease, receive treatment advice, and order the recommended pesticide from a nearby cooperative—all in a few taps. The approach leverages existing mobile networks and minimal cloud processing, making it scalable across rural regions that lack high‑speed internet.
Banga argued that such tools can unlock gains in farming efficiency, improve school attendance through better remote learning resources, and enhance primary health services. The World Bank is exploring additional pilots in Southeast Asia and Sub‑Saharan Africa to extend these benefits.
He also warned that the current hype around large language models obscures the practical needs of low‑resource markets. The five key inputs for running those models—high‑performance computing, massive energy consumption, extensive datasets, skilled personnel, and deep capital—are rarely available together in emerging economies. Consequently, the near‑term opportunity lies with lightweight AI that can be deployed today.