Google has expanded Earth AI, a collection of geospatial AI models and reasoning tools designed to analyse real-world imagery, population and environmental data. The update adds more access to Remote Sensing Foundations and global Population Dynamics Foundations, while riverine flood forecasting now covers over 2B people in 150 countries.
The company said Earth AI is a Google-wide program that builds on its work in floods, wildfires, cyclones, air quality, pollen, weather nowcasting, long range forecasting, agriculture, population dynamics, AlphaEarth Foundations and mobility. Google said these models are already helping millions of people worldwide.
Google said its latest research updates integrate and synthesise data from imagery, population and environmental sources. The Earth AI geospatial reasoning agent uses LLMs and reasoning capabilities to understand concepts and find correlations across multiple datasets and models.
The company said the agent lets users ask complex questions and get answers in plain language, which may make the tools more accessible to non-experts. Google said this could be useful for business logic use cases, supply chain management, crisis resilience and international policy.
In evaluations, Google said the Geospatial Reasoning Agent improved responses over baseline models that did not have access to Earth AI models and tools.
Google said Google Earth with Gemini capabilities will soon be powered by Earth AI imagery models, allowing users to search for objects in satellite imagery. The models are also available to trusted testers on Google Cloud.
The company said partners using the technology include Give Directly, McGill and Partners, Cooper/Smith, WPP, WHO AFRO, and Planet Labs and Airbus.
Source: research.google.
Companies can share verified announcements through Newz9’s international press release submission page.
