According to Google, the newest advancement—called Geospatial Reasoning—lets AI automatically connect different Earth AI models, such as population maps, weather forecasts, and satellite imagery. This enables users to assess where disasters are most likely to strike and who will be most affected.
For instance, GiveDirectly, a nonprofit known for direct aid programs, is using the feature to combine flood and population density data to determine which communities need help first.
“By connecting multiple Earth AI models, Gemini helps organizations move from reacting to anticipating,” said Yael Maguire, Vice President of Geo Sustainability at Google, in the company’s October 2025 announcement.
The integration of Gemini into Google Earth brings new analytical depth. Users can now ask natural-language questions—like “find algae blooms in rivers in the U.S.”—and receive visual insights based on satellite data. This helps analysts identify environmental hazards, such as drought-related dust risks or harmful algae threatening drinking water sources.
The feature will roll out to Google Earth Professional users in the United States in the coming weeks, while AI Pro and Ultra subscribers gain early access with extended limits.
Beyond disaster prediction, Earth AI is already supporting thousands of organizations. The World Health Organization Regional Office for Africa is applying the technology to predict cholera outbreak risks in the Democratic Republic of Congo, combining Earth AI’s population and environmental models with public health data.
Satellite imagery leaders Planet and Airbus are using Earth AI to monitor deforestation and vegetation encroachment near power lines, while Bellwether, an Alphabet X project, is partnering with McGill and Partners to use Earth AI for hurricane prediction and faster insurance claim processing.
The expansion positions Earth AI as a cornerstone of AI-for-climate innovation, blending machine learning with geospatial intelligence to improve both environmental forecasting and resilience planning.
According to UNDRR data, climate-related disasters have increased fivefold over the past 50 years, highlighting the need for advanced predictive tools that can translate complex data into actionable insights. Google’s approach signals a growing convergence between AI reasoning models and climate analytics, giving cities, companies, and nonprofits new ways to anticipate risks rather than react to them.
