Uploaded October 2026 | Updated November 2026, 3 days ago
Public health teams often find themselves fighting today's crises with yesterday’s maps. By complementing existing health data with Google Earth AI’s planetary intelligence, public health leaders can bridge reporting lags, uncover geographic blind spots, and shift emergency response from reactive management to proactive prevention.
Researchers at the World Health Organization’s Regional Office for Africa (WHO AFRO) Emergency Preparedness and Response Hub in Dakar partnered with Google to pilot an experimental Geospatial Reasoning agent during the Bundibugyo strain Ebola outbreak in the Democratic Republic of Congo (DRC).
The WHO AFRO team used Earth AI to map remote mining corridors where ecological exposure intersects with high human mobility. In minutes, the team pinpointed 48 exposed settlements and located more than 45,500 at-risk people—a process that normally takes weeks. This allowed local responders to proactively deploy mobile laboratories and coordinate border surveillance.
Earth AI is a cross-Google program to transform planetary information into actionable intelligence.
Learn more about Google Earth AI → https://ai.google/earth-ai/
Read more about how global public health partners use Google Earth AI → https://blog.google/innovation-and-ai/technology/health/google-earth-ai/
See more Google Research films → https://goo.gle/3NvR4kt
Subscribe to the Google Research Channel → https://goo.gle/GoogleResearch
Public health teams often find themselves fighting today's crises with yesterday’s maps. By complementing existing health data with Google Earth AI’s planetary intelligence, public health leaders can bridge reporting lags, uncover geographic blind spots, and shift emergency response from reactive management to proactive prevention.
Researchers at the World Health Organization’s Regional Office for Africa (WHO AFRO) Emergency Preparedness and Response Hub in Dakar partnered with Google to pilot an experimental Geospatial Reasoning agent during the Bundibugyo strain Ebola outbreak in the Democratic Republic of Congo (DRC).
The WHO AFRO team used Earth AI to map remote mining corridors where ecological exposure intersects with high human mobility. In minutes, the team pinpointed 48 exposed settlements and located more than 45,500 at-risk people—a process that normally takes weeks. This allowed local responders to proactively deploy mobile laboratories and coordinate border surveillance.
Earth AI is a cross-Google program to transform planetary information into actionable intelligence.
Learn more about Google Earth AI → https://ai.google/earth-ai/
Read more about how global public health partners use Google Earth AI → https://blog.google/innovation-and-ai/technology/health/google-earth-ai/
See more Google Research films → https://goo.gle/3NvR4kt
Subscribe to the Google Research Channel → https://goo.gle/GoogleResearch
![Before Rivers Rise | Using AI to Make Global Flood Forecasts
Floods impact millions of people worldwide every year, causing devastation to homes, communities, and livelihoods. But what if we could predict them earlier and more accurately?
Since 2018, Google Research has been building AI technology that helps scale flood forecasting to parts of the world that are most impacted. The system processes vast amounts of data to predict when and where floods will happen, including in regions where data is scarce, empowering local authorities and residents with critical information.
How Google Research’s AI models provide breakthroughs in flood forecasting:
Our leading AI model enhances flood forecast accuracy with 7-day lead time reliability:
https://arxiv.org/pdf/2307.16104
This system powers the Flood Hub ( https://g.co/floodhub ), a platform that provides real time forecasts and alerts. Today, this information can be used directly by people on the ground, communities, governments and aid organizations to take anticipatory action to help protect vulnerable populations.
Increasing reach of our Google-initiated alerts for users: Flood Hub currently covers over 100 countries worldwide for users, providing critical flood forecasting for over 1,800 sites and, reaching over 700 people.
Expanding access to information for experts: To help partners integrate relevant data into their disaster relief response systems and processes, Google Research has also made this global model fully available to governments, partners and researchers, via an API ( https://developers.google.com/flood-forecasting ) that provides real-time 7-day forecasts. We’ve also shared our model forecasts in our Flood Hub mode that leverages virtual gauges intended for use by experts and researchers, and made historical datasets available for use.
What’s next?
Floods are increasingly occurring in unexpected areas. The aim is to leverage the research to expand coverage and improve forecasting of various flood-related events, including flash and urban floods. Additionally, the use of AI to tackle broader climate adaptation challenges and contribute to sustainability efforts.
🔍 Learn more about Googles flood forecasting project: https://g.co/floodhub
Learn more in our latest blog posts: [permalink]
Watch this video in Portuguese: https://youtu.be/Fp1Co2yADJA
This video is perfect for anyone interested in:
AI and climate applications
Al developers
NGO leads working in climate related initiatives
#GoogleFloodForecasting #ClimateTech #AIforGood #Floods #crisisresponse #DisasterPreparedness #globalfloodforecasting Before Rivers Rise | Using AI to Make Global Flood Forecasts](https://i.ytimg.com/vi/iZmltPAFhY0/mqdefault.jpg)









