Michael Kremer says AI could help farmers, governments and public services in poorer countries, with India’s AI-powered monsoon forecasts offering an early example.
Michael Kremer used his first public appearance as the World Bank Group’s chief economist to argue that artificial intelligence could become a powerful development tool for poorer countries, not just a driver of disruption in wealthy ones. The World Bank has centred this year’s development report on the technology, underscoring how strongly its leadership believes AI could influence growth, public services and state capacity in emerging economies.
Speaking at the IMF-World Bank annual meetings in Bangkok, Kremer said AI could make advanced software affordable for small farmers, small businesses and public agencies that have traditionally been priced out of such tools. In his view, the same technology that helps large companies automate and analyse data could also be adapted to improve crop yields, speed up administration and widen access to health care and education in lower-income countries.
He pointed to India as an early example of what that might look like in practice. According to accounts of his remarks, the Indian government has already sent AI-based monsoon forecasts to more than 50 million farmers, helping them decide when to plant and fertilise as weather patterns become less predictable because of climate change. Early studies cited by development observers suggest that plain-language voice messages, delivered in forms farmers can easily use, may lift incomes by making forecasts more actionable.
Kremer’s message reflected a broader divide in how AI is being viewed around the world. In richer economies, policymakers are increasingly focused on the risk that automation will weaken labour markets and erode tax receipts tied to work, while development officials see the technology as a means of strengthening government itself. Ioana Marinescu, an economist at the University of Pennsylvania, has argued that poorer countries rely more heavily on consumption taxes and have fewer jobs exposed to AI disruption, while Miranda Stewart of New York University said tax authorities already use algorithms and could benefit from far better targeting of audits and enforcement.
For Kremer, the central challenge is not simply creating strong models but getting them into everyday use. He has argued that governments will need open-access systems, clear quality standards, public financing and close integration with human services if AI is to improve farm incomes, health care and schooling at scale. That, he suggests, is the point at which AI stops being a promise and starts becoming an instrument of development policy.
Disclaimer: This article is intended to inform and educate, not to recommend or endorse any financial product, investment or strategy. Please consider your own financial circumstances and seek professional advice where appropriate before making financial decisions.
