IBM and NASA Release Groundbreaking Open-Source AI Model on Hugging Face to Predict Solar Weather and Help Protect Critical Technology
The release of Surya, an open-source AI foundation model developed by IBM and NASA, represents a significant leap in solar weather forecasting. Trained on high-resolution solar observation data, this model addresses the critical vulnerability of modern technology to space weather, which can disrupt power grids, satellites, and navigation systems. By providing more accurate predictions of solar activity, Surya helps mitigate economic and operational risks associated with solar storms, offering a protective layer for infrastructure that humanity increasingly depends on. This advancement relies on the largest curated heliophysics dataset ever released, enabling unprecedented spatial resolution in analyzing the Sun’s dynamic surface. Unlike previous methods that struggled with partial views, Surya can visually predict solar flares up to two hours in advance, substantially improving classification accuracy. This capability transforms how scientists study solar behavior, moving beyond binary classification to detailed visual forecasting, thereby enhancing our understanding of the complex interactions between solar activity and Earth’s technological environment. For the open data community, this initiative is highly relevant as it democratizes access to advanced scientific tools through open-source platforms like Hugging Face. By releasing both the model and the underlying dataset, IBM and NASA empower global researchers to build upon existing work, fostering collaborative innovation. This approach accelerates scientific discovery and ensures that critical space weather prediction technologies are accessible to the broader research and technology community, rather than being confined to a few institutions.
Source: manilatimes.netPublished on 2025-08-21