As the climate is changing, an increasing occurrence of extreme events combined with random shift in seasonal weather patterns is leading to a high uncertainty in food supply. In this research, we present a methodology that combines seasonal forecasts of weather and extreme events, along with agronomic data using AI methods to predict risk to food production at scale. In our case study, we forecast supply of corn across a large state in India, with lead times of up to 4 months. Our method can predict other important considerations such as harvest window prediction and water footprint. Our analysis shows the potential of AI and geo-spatial data analytics to better quantify future food production.