{
  "id": 9628142,
  "title": "AI/ML News",
  "url": "https://urgent.news/2026/09/24/ai-ml-news",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-24T20:06:52.000Z",
  "source": {
    "name": "NASA Science",
    "slug": "nasa-science",
    "url": "https://www.nasa.gov/ai-astrobiology/ai-ml-news/"
  },
  "original_language": "en",
  "account": "NASA's AI/ML research has been making significant strides, with a particular focus on foundation models (FMs) for various scientific applications, including lunar and Martian exploration. In 2023, the NASA Interagency Implementation and Advanced Concepts Team (IMPACT) and IBM collaborated to develop a geospatial FM called Prithvi, trained on Landsat and Sentinel-2 data. Prithvi, also known as \"the Moon and Beyond,\" can be used for fire monitoring, flood analysis, and crop yield estimation. This led to the creation of a \"5+1\" strategy from NASA's Office of the Chief Science Data Officer (OCSDO), envisioning FMs for each science division in the Science Mission Directorate, along with a language model to facilitate user access. The Planetary Science Division has initiated its own FM projects, such as the Lunar FM and Mars FMs, which will utilize lessons learned from previous FM efforts and scientific input from across NASA and the community. Recently, the open-source Lunar FM was released on Hugging Face and GitHub, enabling scientists to map craters, detect active volcanic features, estimate ice locations, and potentially aid future lunar missions. The next steps involve developing two Mars FMs—one for the planetary surface and one for the atmosphere—and integrating the NASA Science Discovery Engine's INDUS suite of science-focused language models to enhance user search and discovery capabilities.",
  "summary": "Update on NASA Foundation Model Progress: To the Moon and Beyond September 2026 – NASA’s foundation model (FM) work began taking shape in 2023 when the NASA Interagency Implementation and Advanced Concepts Team (IMPACT) and IBM formed a collaboration to build a geospatial FM. Due to the Earth being so well studied and mapped out […] The post AI/ML News appeared first on NASA Science .",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}