{"id":536729,"date":"2026-09-10T22:00:00","date_gmt":"2026-09-10T12:00:00","guid":{"rendered":"https:\/\/science.nasa.gov\/science-research\/artificial-intelligence-lunar-foundation-model\/"},"modified":"2026-09-10T22:00:00","modified_gmt":"2026-09-10T12:00:00","slug":"nasa-ibm-launch-ai-foundation-model-for-lunar-science","status":"publish","type":"post","link":"https:\/\/www.vibewire.com.au\/?p=536729","title":{"rendered":"NASA, IBM Launch AI Foundation Model for Lunar Science"},"content":{"rendered":"
\n
\n
\n

6 min read<\/p>\n

NASA, IBM Launch AI Foundation Model for Lunar Science<\/h1>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
\"Overhead<\/a><\/figure>
\n
A 10-image mosaic captured by NASA\u2019s Lunar Reconnaissance Orbiter\u2019s Narrow Angle Camera between June 2012 and April 2016 showing the volcanic feature Mons R\u00fcmker and its surrounding mare plains.<\/div>\n
NASA\/GSFC\/Arizona State University<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

NASA is bringing artificial intelligence to the study of the Moon, helping researchers transform how they analyze the Moon\u2019s surface. In an ongoing collaboration with IBM Research and several academic institutions, NASA has launched the NASA-IBM Lunar Foundation Model, among the first open-source AI models built specifically for lunar science. The model, trained primarily on data from NASA\u2019s Lunar Reconnaissance Orbiter (LRO), is hosted publicly on Hugging Face<\/a> for anyone to use, with the complete codebase available on GitHub<\/a> for testing and experimentation.<\/p>\n

The NASA-IBM Lunar Foundation Model supports the next generation of lunar science by helping researchers quickly analyze vast quantities of data to better understand the Moon\u2019s surface. Using the model as a mapping tool, researchers can rapidly develop actionable strategies for evaluating the Moon\u2019s rugged surface, understanding its geological past, and planning future lunar research.<\/p>\n

\u201cNASA has spent decades building an extraordinary scientific record of the Moon, but collecting data is only part of the job,\u201d said Kevin Murphy, chief science data officer and acting chief data and AI officer at NASA Headquarters in Washington. \u201cWe also have to make data easier for scientists to explore and use. The NASA-IBM Lunar Foundation Model shows what\u2019s possible when we bring AI to NASA\u2019s petabytes of scientific data. That\u2019s a real opportunity we see with AI: turning large-scale data into new discoveries.\u201d<\/p>\n

Unlike traditional models that require building and training specialized algorithms from scratch for specific tasks, foundation models are pre-trained on vast, unlabeled datasets. The broad knowledge they acquire through pre-training allows them to generalize across multiple scientific domains through quick fine-tuning, making foundation models both versatile and efficient in accelerating scientific research.<\/p>\n

\n
\n
\n
\n\t\t\t\t <\/path><\/path><\/g><\/svg>\n\t\t\t<\/div>\n
\n
\n

The NASA-IBM Lunar Foundation\u00a0Model shows\u00a0what\u2019s\u00a0possible when we bring AI to NASA\u2019s petabytes of scientific data.<\/span><\/h2>\n<\/p>\n<\/div>\n
\n
\n

Kevin Murphy<\/p>\n

NASA Chief Science Data Officer and Acting Chief Data Officer\/Chief AI Officer<\/p>\n<\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n

Data collected by NASA\u2019s LRO over the past 17 years was well-suited for training this foundation model because it covers most of the lunar surface in detail. The data produced from the LRO mission is larger than all other NASA planetary missions combined, capturing an almost seamless, high-resolution mosaic of the entire Moon. The NASA-IBM model was trained on roughly 2 million image tiles from this dataset, comprising more than 1 million high-resolution camera images at 1-meter resolution and nearly 964,000 multispectral images at 100-meter resolution. The model also was trained on high-resolution Moon imagery and terrain data from multiple other missions such as NASA\u2019s GRAIL<\/a> (Gravity Recovery and Interior Laboratory), NASA\u2019s Lunar Prospector,<\/a> and JAXA\u2019s (Japan Aerospace Exploration Agency) Selenological and Engineering Explorer.<\/p>\n

Because the foundation model is already pre-trained on this dataset, planetary scientists can adapt the model to many different lunar research tasks such as mapping craters, spotting young volcanic features, and estimating where ice may exist near the lunar poles by using only small amounts of labeled data. For researchers who study the Moon\u2019s polar ice, the NASA-IBM model can help them estimate where ice patches are likely to be stable, on and below the surface. Dark areas like the Moon\u2019s permanently shadowed regions remain cold enough to trap and preserve ice for up to billions of years. Studying these areas offers insight into the Moon\u2019s history and presents an opportunity to map potentially usable resources for future space exploration.<\/p>\n

\n
\n
\n
\"The<\/a><\/figure>
\n
The NASA-IBM model reproduces patterns of lunar ice prospectivity (scaled from blue to yellow), shown at four locations (left) near the Moon\u2019s pole. Top row: reference ice prospectivity map of Mons Mouton near the lunar south pole; middle row: predictions from the ConvNeXt model; bottom row: predictions from the NASA-IBM model. The NASA-IBM model preserves many fine-scale prospectivity patterns in the reference data.<\/div>\n
NASA\/IBM Research<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

While the Moon is thought to no longer be volcanically active, it once experienced dynamic geological processes. For researchers studying lunar volcanism, the NASA-IBM model accelerates the identification of unusual looking volcanic features known as irregular mare patches. Because these structures appear relatively young, they challenge established timelines for lunar cooling, and mapping them could help scientists piece together a more accurate understanding of the Moon\u2019s thermal evolution.<\/p>\n

The model also can map surface features, such as craters, more efficiently than manual methods. Every crater is formed by an impact, making crater counts and measurements essential for dating the lunar surface and reconstructing solar system history. The foundation model helps speed up the process of identifying and measuring craters, allowing scientists to focus on interpreting findings and determining their implications for exploration.<\/p>\n

\n
\n
\n
\"Side-by-side<\/a><\/figure>
\n
These Lunar Reconnaissance Orbiter images show the Moon\u2019s surface near Einstein crater before (left) and after (right) a SpaceX rocket body impact. The NASA-IBM Lunar Foundation Model detected existing craters (blue outlines) and highlighted the newly formed impact crater (red box). Because the post-impact image was excluded from pre-training, this test demonstrates how the model can be fine-tuned to recognize novel surface changes between observations. This approach can help scientists automatically detect natural impacts and surface changes across vast lunar datasets, though varying lighting conditions between orbits may influence smaller crater visibility.<\/div>\n
NASA\/IBM Research<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

Overall, the model matched or exceeded the performance of several other strong baseline models across all evaluated tasks, achieving comparable results on crater mapping and segmentation of irregular mare patches, while demonstrating a clear advantage on estimating polar ice stability.<\/p>\n

The NASA-IBM Lunar Foundation Model is part of the agency\u2019s Office of the Chief Science Data Officer\u2019s strategy for AI for science \u2014 a larger, ongoing collaboration between NASA and IBM aimed at using advanced AI to explore our planet and solar system. It joins a growing collection of AI models developed through this partnership, including:<\/p>\n