We’re excited to share the latest case study in our ongoing series on how artificial intelligence (AI) is transforming humanitarian response. This edition highlights GiveDirectly’s innovative use of AI-powered flood forecasts to deliver anticipatory cash assistance in Nigeria and Bangladesh.
With support from Google.org and technology developed by Google Research, GiveDirectly has piloted a system that predicts flood events at the village level, triggering early cash payments to help vulnerable communities prepare before disasters strike. By combining advanced AI forecasting with locally sourced data and on-the-ground expertise, the initiative is setting a new standard for precision, speed, and impact in humanitarian aid.
The case study, published in March 2025, explores how GiveDirectly, in partnership with JBA Global Resilience, operationalized these AI-generated triggers to deliver cash within 48 hours—empowering communities to evacuate livestock, fortify homes, and protect livelihoods. It also shares valuable lessons about the importance of local context and continuous calibration to ensure forecasts reflect real-world conditions.
We’re proud to feature this compelling example of technology and local knowledge working hand-in-hand to improve disaster preparedness and preserve dignity. Special thanks to Vera Lummis and Federico Barreras at GiveDirectly for their contributions to this important work.
If you are a NetHope Member, sign up to the NetHope Network here and to join the NetHope AI and Emerging Technologies Working Group to learn and collaborate on the responsible use of AI for nonprofits.
