Earth Action in Transition: Highlights From the 2025 ESA–NASA International Workshop on AI Foundation Models for EO [Space-Agencies]
Authors:
- Nicolas LongépéNLNicolas Longépé
- Hamed AlemohammadHAHamed Alemohammad
- Anca AngheleaAAAnca Anghelea
- Thomas BrunschwilerTBThomas Brunschwiler
- Gustau Camps-VallsGCGustau Camps-Valls
- Gabriele CavallaroGCGabriele Cavallaro
- Jocelyn ChanussotJocelyn Chanussot
- Begüm DemirBegüm Demir
- Nikolaos DionelisNDNikolaos Dionelis
- 17 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Magazine
- Vol. 13 (4) , 457-462
Abstract
Over 850 people joined the first International Workshop on AI Foundation Model (FM) for Earth Observation (EO) co-organized by ESA and NASA 5-7 May 2025. Hosted at ESRIN (ESA’s Earth Observation Center, Italy), the event welcomed around 300 people on site, and an additional 550 online, with the promise that FMs can revolutionize EO and Earth sciences. The workshop marked a pivotal moment in aligning the EO and FM communities, fostering a shared commitment to developing open and trustworthy tools that support science discovery, operational applications, and prescriptive analytics. EO data is massive, complex and high dimensional requiring specific yet scalable AI architectures. The workshop emphasized the need for training and architecture enabling interpretability, explainability, and physical consistency. Coordination should be strengthened to minimize redundant development and to better leverage collective expertise. The focus has shifted from prototyping to real-world deployment, with FMs needing further design for integration into digital twins, dashboards, and edge platforms. Transparent benchmarking and user-driven evaluation are key to guiding model development and decision-making. In addition, parameter-efficient adaptation, neural compression, and embedding-based workflows offer promising paths for scaling EO analytics. While FMs show promise, their effectiveness remains context-dependent. The community debated whether to pursue universal models, specialized solutions, or mixtures of experts. The workshop envisioned the future of agentic AI in EO, with multi-agent system powered by EO FMs and vision-language models, that can dynamically reason and act on EO data. This shift from static pipelines to adaptive, smarter systems could redefine the future of EO. This paper summarizes key discussions and concludes with thought-provoking remarks.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- Not Every Tree is A Forest: Benchmarking Forest Types from Satellite Remote Sensing2025
- Towards Efficient Benchmarking of Foundation Models in Remote Sensing: A Capabilities Encoding Approach2025
- SARFormer - An Acquisition Parameter Aware Vision Transformer for Synthetic Aperture Radar Data2025
- AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalities2025
- SenPa-MAE: Sensor Parameter Aware Masked Autoencoder for Multi-satellite Self-supervised Pretraining2025
- Fine-Tuning Foundation Models With Confidence Assessment for Enhanced Semantic SegmentationIEEE Geoscience and Remote Sensing Letters, 2024
- MMEarth: Exploring Multi-modal Pretext Tasks for Geospatial Representation Learning2024
- OmniSat: Self-supervised Modality Fusion for Earth Observation2024
- PhilEO Bench: Evaluating Geo-Spatial Foundation Models2024
- MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers2023