NASA and IBM release open-source AI model for lunar exploration

NASA and IBM have released the NASA-IBM Lunar Foundation Model, an open-source artificial intelligence tool designed to accelerate lunar scientific exploration. The model is trained on decades of observations, utilizing over 30 data layers from nine instruments across four NASA missions, including the Lunar Reconnaissance Orbiter.

In benchmark testing, the model demonstrated up to 23% greater accuracy than existing methods in identifying key geographic features. Its primary applications include locating potential water ice deposits in permanently shadowed regions, mapping craters to identify safe landing sites, and studying volcanic formations known as Irregular Mare Patches.

The release supports NASA’s Artemis program, which aims to establish a sustained human presence on the Moon. Identifying resources like ice is critical, as it provides water, oxygen, and potential rocket propellant for future missions to Mars. The model is part of IBM’s Prithvi family of scientific foundation models and is available on platforms such as Hugging Face and GitHub.