Building the open infrastructure needed for GeoAI to scale globally
Taylor Geospatial advances the state of the art in GeoAI to address global challenges through partnership and innovation. Our work democratizes access to information held in satellite imagery at global scale.
Select a program to explore
Food Systems Challenge
The Geospatial Innovation for Food Security (GIFS) Challenge promotes advanced research in three critical food systems problem spaces: enabling agri-food supply chain resilience, informing crop shifting, and increasing nitrogen use efficiency.
Benchmark Dataset
Significantly expands the world's leading open dataset for mapping agricultural field boundaries from satellite imagery — improving geographic coverage, data richness, and the range of tasks models can support.
Global Feature Detection
Applies the field-boundary approach to a broader set of physical features visible from space — roads, buildings, solar panels, trees, and more — training models capable of detecting and mapping them globally.
Evaluation Infrastructure
Creates a shared, community-governed system for evaluating geospatial AI models — assessing real-world performance, failure modes, compute costs, and generalization across geographies.
We have a big vision for how geospatial AI, released as digital public goods, can change the world. We identify and engage innovative researchers and technical fellows before inviting them to submit a proposal for funding. Most Taylor Geospatial grants are made through these invitations.
We look for opportunities to co-invest with organizations that share our vision and values. Our goal is to work with the best in the industry to jointly create a future where geospatial data and AI models are digital public goods with a positive impact on global decision making.
Contact VP for Strategic Innovation, Jen Marcus at jennifer.marcus@taylorgeospatial.org to discuss a potential co-investment partnership.
All benchmarks, training data, models, model weights and code will be released as digital public goods. Fields of The World, Features of The World, and Benchmarks of The World are structured so their outputs are available the moment they're ready — not locked in journals or behind licensing agreements. The goal is an ecosystem where interoperability and usability are the norm, and the power of GeoAI can be deployed everywhere it is needed.