IBM and NASA release AI model mapping Moon ice and craters

IBM and NASA released the Lunar Foundation Model, an open-source artificial intelligence system designed to analyze decades of lunar observation data. The model was trained on more than 30 layers of imagery collected by nine instruments across four NASA missions, including the Lunar Reconnaissance Orbiter. This capability aims to assist scientists in identifying water ice deposits within permanently shadowed regions, mapping crater locations for safe landing sites, and studying volcanic features. Previously, these tasks required researchers to manually sift through maps or rely on lower-resolution machine-learning tools. During benchmark tests, the new model identified surface features 23 percent more accurately than widely used methods. Lunar ice is critical because it represents water and oxygen resources necessary for future manned bases and producing rocket fuel for missions to Mars. The company stated the tool joins the Prithvi family of open foundation models, which covers geospatial and weather applications. NASA plans to return astronauts to the Moon in 2028 using new technology tested during this sustained presence program.