Google DeepMind has launched AlphaEarth Foundations, an advanced AI model described as a "virtual satellite" that integrates petabytes of Earth observation data to create highly detailed, 10-meter resolution global maps. This model processes trillions of disparate datasets, including satellite imagery, to generate compact, consistent summaries of terrestrial land and coastal waters. AlphaEarth Foundations enables near real-time monitoring of critical environmental issues such as food security, deforestation, water resources, climate change, urban development, and ecosystem shifts. The AI model produces analysis-ready data embeddings for each year dating back to 2017 without requiring additional atmospheric corrections or cloud masking. It is designed to support scientists and policymakers in making informed decisions on global challenges like flood and wildfire forecasting, weather prediction, and sustainable urban planning. The technology represents a significant advancement in geospatial AI, offering unprecedented precision and scalability for planetary monitoring and climate action.
🛰️ 🔭 Say hello to Galileo, an open-source multimodal model designed to process many kinds of earth observations at once—driving applications such as mapping agricultural land, detecting floods, and monitoring marine pollution.
AlphaEarth: La nueva IA de Google que promete cambiar para siempre la forma en que vemos nuestro planeta https://t.co/4BegbgNvMl
Artificial Intelligence Finds Holes on the Moon Scientists have adapted deep learning algorithms to search for underground lava tubes and caves on the Moon. Using satellite images, the ESSA model successfully found 2 new entrances to the pits. Moreover, they were discovered https://t.co/sbAldKlAvX