US Edition
Google is using old news reports and AI to predict flash floods
- Thursday, March 12, 2026, at 3:45 PM UTC, Google Research shared Groundsource and published its dataset publicly, highlighting risks for urban areas in 150 countries via Flood Hub.
- Facing scarce local records, researchers sought news data because flash floods kill more than 5,000 people yearly and are hard to predict due to their short-lived, localized nature.
- Google used Gemini to sort through 5,000,000 news articles and isolate 2,600,000 flood events to create the Groundsource time series, then trained a Long Short-Term Memory neural network model.
- Flood Hub currently showcases areas at high flash flood risk in the near future, and trial users, including António José Beleza, said it helped emergency response agencies react faster.
- The model has notable limits, as it identifies risk across 20-square-kilometer areas and lacks local radar data, making it less precise than the US National Weather Service, while accuracy metrics and ground-truth data remain scarce for evaluation.
23 Articles
23 Articles
Google has built a predictor of the future, using the Gemini language model, but millions of press articles. It is for the first time when the company uses large language models for such predictions. The futures are among the most dangerous weather phenomena in the world, causing about 5,000 floods per year. ...
Observador
DnoticiasWith 2.6 million historical flooding recorded in more than 150 countries, the new Google system promises to warn communities up to 24 hours before recent urban floods.
Groundsource uses historical data and Artificial Intelligence to predict potential natural disasters around the world
Google Is Using AI to Fill a Flood Risk Data Gap
Flash floods, when stormwater pools and rises rapidly in an area within just a few hours of a storm's onset, are one of the more dangerous hazards of a warming planet prone to heavier rainfall. They are also notoriously difficult to predict. But research out of Google on Thursday shows how artificial intelligence could unlock better forecasts and help communities prepare. Google researchers used Gemini, the tech giant’s signature AI agent, to pr…
Coverage Details
Bias Distribution
- 72% of the sources lean Left
Factuality
To view factuality data please Upgrade to Premium
