Archive · 10 Sep 2026
NASA and IBM Launch Open-Source AI Model for Moon Mapping
NASA and IBM created a new computer tool to study the Moon.
The tool uses artificial intelligence to examine many years of lunar observations.
It learned from more than 30 kinds of data collected by nine instruments on four NASA missions.
The model can help scientists look for ice in places that are always in shadow.
It can also map craters and help find safer places for spacecraft to land.
Scientists can use it to study old volcanic features on the Moon.
Lunar ice could provide water and oxygen for future astronauts and rocket fuel for Mars missions.
NASA’s Artemis program plans to return astronauts to the Moon in 2028.
NASA and IBM released a publicly available, open-source AI model for analyzing decades of lunar observation data.
The NASA-IBM Lunar Foundation Model was trained on more than 30 data layers from nine instruments on four NASA missions.
The model can help identify potential ice deposits, map craters and study lunar volcanic features.
Researchers could use it to support safer landing-site selection and plans for a sustained human presence on the Moon.
NASA and IBM said the model identified key lunar surface features up to 23% more accurately than widely used methods.
- Who
- NASA and IBM released the NASA-IBM Lunar Foundation Model.
- What
- They released an open-source AI model for analyzing lunar data and mapping features on the Moon.
- When
- The model was released on Thursday; NASA’s Artemis program plans a return to the Moon in 2028.
- Where
- The model is designed to study the Moon, including permanently shadowed regions.
- Why
- It is intended to help identify lunar ice, select safer landing sites, study volcanic features and support sustained human missions.
- Model name
- NASA-IBM Lunar Foundation Model
- Availability
- Publicly available and open source
- Training data
- More than 30 data layers from nine instruments on four NASA missions
- Featured mission
- The Lunar Reconnaissance Orbiter contributed data
- Potential uses
- Ice detection, crater mapping, landing-site selection and volcanic-feature analysis
- Benchmark result
- Up to 23% greater accuracy than widely used methods
- Future mission link
- NASA’s Artemis program plans to return astronauts to the Moon in 2028
