Release of open AI model trained on 17 years of lunar data maps ice, craters and volcanoes

NASA and IBM have released an open-source lunar foundation model pre-trained on 17 years of orbital data, including roughly two million image tiles primarily from the Lunar Reconnaissance Orbiter. By requiring only small amounts of labeled data for fine-tuning, the model accelerates planetary science tasks such as counting impact craters, identifying young volcanic structures, and estimating polar ice stability. This release demonstrates how generalizable foundation models can unlock petabytes of raw planetary remote-sensing data for reproducible research and exploration planning.

Read the original at phys.org   Generate teaching or study materials

Instructors get discussion guides, assignments, and mini-cases. Students and readers get a plain summary, class prep, and an exercise. All built from the full article. Three are free with an account.

More in Astronomy

• AI’s best coding agent fails 60% of the time — and the data backs it upThe New Stack · September 15, 2026
• Observing Earth from orbit is unpredictable and messy: Could 'liquid' AI clear the view?Phys.org — Technology · September 15, 2026