The Institute for Climate and Sustainable Growth’s Human-Centered Weather Forecasts Initiative (HCWF) is collaborating with the Ethiopian Meteorological Institute (EMI) to improve weather forecasts for farmers throughout the country. The team will build AI-driven forecasts tailored to the specific needs of Ethiopian farmers, and work with local meteorologists to operationalize them and build capacity for future improved forecasting. The work, supported by the Gates Foundation, is part of a larger project to integrate AI-driven techniques into weather forecasting throughout several African countries.
“Reliable weather information is the backbone of agricultural resilience, but traditional forecasts have historically required supercomputers and dense weather-observing station networks that many nations lack,” says Katie Kowal, director for AI for Weather at HCWF and its sister organization AI for Climate (A joint initiative of the Institute for Climate and Sustainable Growth and the Data Science Institute). “By providing governments with the training, hardware and institutional capacity to adapt and use AI models, we are driving a paradigm shift in weather forecasting that brings high-quality and actionable information to people living in low-and middle-income countries.”
The work in Ethiopia was launched together with the Agricultural Innovation Mechanism for Scale (AIM for Scale), a global initiative that works to scale up evidenced-backed, cost-effective agricultural innovations for the benefit of farmers throughout the Global South. It also includes the UChicago Development Innovation Lab (DIL) and Precision Development (PxD), a global nonprofit supporting smallholder farmers in digital advisory services. DIL and PxD are designing and testing the forecast messages that will go to farmers. HCWF, AIM for Scale and PxD previously worked together to support the Government of India’s Ministry of Agriculture and Farmers Welfare in distributing AI-based forecasts to 38 million Indian farmers during the 2025 monsoon season. That project is now a global model for how national governments, like in Ethiopia, can use AI-driven forecasts to help farmers adapt to climate change.
Kowal and a team from HCWF met with their collaborators recently for a workshop on February 2, 2026, convening more than 30 government and technical partners to align on next steps for scaling weather forecasts in Ethiopia. A roundtable followed, which brought together government officials from the Ethiopian Agricultural Transformation Institute and the Ethiopian Ministry of Agriculture to align and coordinate ways to scale the work.
Building on this momentum, Kowal presented an AI-model overview to Ethiopian meteorologists on February 4. This technical exchange was part of the build-up to the National Climate Outlook Forum on February 9—an event dedicated to enhancing resilience through seasonal information services in Ethiopia—which Kowal also attended along with HCWF team members Bo Dong, a UChicago senior computational scientist and Giulio Schinaia, from DIL. This provided HCWF researchers with the unique opportunity to learn more about the seasonal forecast processes and cross-sector feedback mechanisms led by EMI and its partners.
The inaugural phase of this new project in Africa closed with technical sessions on February 11 and 12 led by Dong. HCWF collaborated with EMI colleagues to examine what would signal the start of the rainy season and began to build this knowledge into the AI model development process.
“The Ethiopian Meteorological Institute is working with AIM for Scale and the University of Chicago to ensure that climate information leads to better, more actionable decisions, leveraging new AI tools,” said Dr. Fetene Teshome, Director General of EMI. He underscored the importance of strengthening the link between artificial intelligence and real-world applications. “Climate variability is projected to continue increasing. This is likely to result in more irregular onset and cessation of the rainy season, longer dry spells, and periods of excessive rainfall. As a result, farming systems and other sectors of the economy are becoming increasingly vulnerable. The role of climate information in minimizing risks and optimizing opportunities is therefore non-debatable.”
Images for Sanity website - 1
Kowal
Images for Sanity website - 1
