By Yossi Matias
Publication Date: 2026-03-12 00:00:00
When disaster strikes, information is a lifeline. As part of Google’s Crisis Resilience efforts, we have been providing early warning of natural hazards for years to help communities stay safe. However, for certain disasters such as flash floods, there simply was no reliable data. This data gap has long prevented us from training AI models to predict flash floods before they happen – until now.
Today we’re introducing Groundsource, a new AI-powered method that transforms public information into a high-quality record of historical disaster data – starting with flash floods in urban areas. Groundsource used Gemini to analyze decades of public reports and identified over 2.6 million historical flood events in more than 150 countries. Google Maps was then used to determine precise geographical boundaries for each event and create a data set focusing on flash floods. Using this dataset, we trained a new model that makes noticeable progress in predicting flash floods…
