Famine Forecasting: Who Benefits, Who Is Left Behind?

This project critically examines the growing use of artificial intelligence (AI) tools to forecast food insecurity in fragile and conflict-affected settings. Humanitarian organizations such as the World Food Programme (WFP) and FEWS NET increasingly rely on predictive models to anticipate famine risk and guide resource allocation. However, these systems often depend on incomplete, biased, or exclusionary datasets that may fail to account for the needs of displaced communities, informal food systems, and under-documented regions. This research explores how data gaps and algorithmic assumptions shape humanitarian decision-making, drawing on case studies from Yemen, South Sudan, and Syria. The final report will assess the effectiveness and ethical implications of AI-based famine forecasting tools and offer recommendations for more inclusive and accountable systems.