NASA and IBM Release Open-Sourced Lunar AI Model

IBM and NASA have released an open-source AI model called the NASA-IBM Lunar Foundation Model, aimed at supporting a sustained human presence on the Moon. The model is designed to integrate observations from multiple instruments and missions, allowing researchers to query and fine-tune it for specific tasks.

The model was trained on data from nine instruments across four NASA missions, including the Lunar Reconnaissance Orbiter (LRO), which has been mapping the Moon since 2009. In benchmark testing, the model identified key lunar surface features up to 23% more accurately than prior methods.

The model's developers, IBM Research and NASA, point to three concrete use cases: identifying potential ice deposits inside permanently shadowed regions, mapping craters at scale for safer landing sites, and studying volcanic features on the lunar surface. The model is distributed as an open-source release, with trained weights hosted on Hugging Face and supporting code published on GitHub.