Google announced new progress in its decade-long effort to use AI and global partnerships to help communities prepare for and respond to natural disasters.
The company said its crisis resilience work is guided by a goal of moving toward a world where no one is surprised by a natural disaster. Google has been developing AI-based research and tools for forecasting, detecting, and sharing timely information about floods, wildfires, earthquakes, extreme weather, heat, and air quality.
Google’s flood forecasting work began with a pilot in the Patna region of India in 2018. Since then, the company has expanded its flood forecasting models globally, including to data-scarce regions.
Google said forecasts on Flood Hub now cover 2 billion people across more than 150 countries in areas at risk for significant flood events. River flood forecasts are available up to 7 days in advance, while new flash-flood predictions in urban areas can provide up to 24 hours of advance notice.
The company also introduced Groundsource, an AI-based methodology that created a high-quality flood dataset from 20 years of public reports. Google said it used that dataset to train a flash-flood model and has open-sourced both the flash-flood dataset and its hydrology framework.
For extreme weather, Google highlighted WeatherNext 2, which it said delivers its most accurate predictions yet. The model can generate detailed hourly global forecasts in minutes and predict variables including wind speed, wind direction, precipitation, and pressure. During the 2025 hurricane season, Google said WeatherNext 2 successfully predicted the path and intensity of cyclones with high confidence days in advance.
Google is also using satellite imagery to track wildfire boundaries using AI in Search and Maps. The company said it has expanded wildfire coverage to 34 countries, including seven new countries this year.
To improve wildfire detection, Google co-developed FireSat in collaboration with the Earth Fire Alliance and Muon Space, with support from Google.org, the Moore Foundation, the Bezos Earth Fund, and others. The first protoflight satellite was placed in orbit last year, and a full FireSat constellation of more than 50 satellites would be designed to detect wildfires as small as 5 x 5 meters anywhere on Earth, with updates every 20 minutes.
Google is also applying AI to extreme heat. The company is using satellite and aerial imagery to map building reflectivity across urban environments, which could help cities identify opportunities to reduce surface temperatures through cool roofs.
The company has also brought climate and geospatial models together in the Google Earth AI collection of models and datasets. Google said the collection supports planetary intelligence and can help organizations address complex challenges such as disaster response and planetary monitoring.
Google continues to provide crisis updates across Search and Maps through SOS alerts, which bring together information from authorities and trusted media outlets. The company also works with authorized alert originators and distributors in more than 90 countries to amplify emergency alerts and public warnings through Public Alerts.
Google said its crisis information received billions of views, and last year it helped connect people with crisis information more than 10 million times per day on average.
Other crisis-related tools include extreme heat alerts on Search for people in more than 100 countries, Android Earthquake Alerts that can warn users before shaking reaches them, and air quality information on Google Maps in more than 30 countries.
Google said collaboration remains central to building global resilience. The company works with governments, UN agencies, organizations, scientists, and first responders to help communities prepare for disasters.
In Nigeria and Bangladesh, GiveDirectly and the International Rescue Committee have used Google’s flood forecasts to support anticipatory action, including distributing emergency cash before rising waters so communities could evacuate and protect belongings. Google also said the U.S. National Hurricane Center used its WeatherNext model during Hurricane Melissa, which predicted Jamaican landfall five days in advance, helping the Meteorological Service of Jamaica notify the public.
KEY QUOTES:
“The world is experiencing a dramatic rise in extreme weather events and natural disasters, devastating communities. Over the past decade, our teams at Google have worked to make helpful information available to people at times of crises — often when they need it most.”
“Actionable information in times of crises can help save lives and livelihoods: our north star for our crisis resilience efforts is that no one should be surprised by a natural disaster.”
“Over the past decade, we’ve made progress driving AI-based research breakthroughs and solutions for climate resilience, providing actionable, timely information to communities around the world. I’m optimistic that by harnessing AI and working with our partners, we’ll move closer towards a world where no one is surprised by a natural disaster.”
Yossi Matias, VP, GM of Google Research and Google Crisis Resilience Lead
