NASA And IBM Are Developing AI Model To Monitor The State Of The Earth

NASA estimates that the Earth science program will generate about a quarter of a million terabytes of data in 2024. To enable climatologists and the research community to efficiently process these raw satellite data sets, IBM, HuggingFace and NASA will create an open source geospatial database that will serve as the basis for a new class of artificial intelligence that can track deforestation, predict crop yields and more.

To enable climatologists and the research community to efficiently process these raw satellite datasets, IBM, HuggingFace and NASA are collaborating to create an open source geospatial model that will serve as the foundation for a new class of climate and earth science artificial intelligence that  will be able to track deforestation, forecast crop yields and estimate greenhouse gas emissions.

For this project, IBM used the recently released Watsonx.ai as a reference model using data from the NASA Harmonized Landsat Sentinel-2 (HLS) satellite, which is collected by ESA and serves as the basis for creating AI.

In turn, HuggingFace hosts the model on its open source artificial intelligence platform.