IBM and NASA Release Open Source Geospatial AI Foundation Model 

Landsat imagery captured in June 2023 shows the shrinking Paso Severino reservoir in Uruguay. Photo: NASA

IBM announced this week that IBM’s geospatial foundation model watsonx.ai, which is built from NASA‘s satellite data, will now be openly available on AI platform Hugging Face. IBM says this will be the largest geospatial foundation model on Hugging Face and the first-ever open-source AI foundation model built in collaboration with NASA.

The model was trained jointly by IBM and NASA on Harmonized Landsat Sentinel-2 satellite data (HLS). The model was trained for data for flood and burn scar mapping and can also be redeployed for tasks like tracking deforestation, predicting crop yields, or detecting and monitoring greenhouse gasses.

NASA and IBM said that making it available on Hugging Face will democratize access and lead to innovations in climate and Earth science. This comes after NASA and IBM started a new agreement earlier this year to build an AI foundation model for geospatial data. 

“The essential role of open-source technologies to accelerate critical areas of discovery such as climate change has never been clearer,” said Sriram Raghavan, vice president of IBM Research AI. “By combining IBM’s foundation model efforts aimed at creating flexible, reusable AI systems with NASA’s repository of Earth satellite data, and making it available on the leading open-source AI platform, Hugging Face, we can leverage the power of collaboration to implement faster and more impactful solutions that will improve our planet.”