get in touch
TGI/AWS Generative AI for Geospatial Challenge
Generative Artificial Intelligence (GenAI) is rapidly reshaping how we understand, interact with, and act upon data. Nowhere is this transformation more promising — or more necessary — than in the geospatial domain. Explore this page to learn more.
Download Full report
challenge sponsors
TGI is bringing together leading geospatial researchers and some of the largest geospatial data providers to push the boundaries of what’s possible in making the planet's data more accessible and queryable. With the broadest and most comprehensive set of cloud services, AWS is at the forefront of this transformation, empowering customers to accelerate the development of highly scalable, cloud-native geospatial solutions. The Challenge is a key initiative in the broader AWS and TGI collaboration, designed to inspire and support the creation, commercialization, and scaling of innovative geospatial applications.

To fuel this innovation, AWS offered $1,000,000 in AWS credits, providing participants with the resources to think big and turn visionary ideas into reality.
Acknowledgements
At the Taylor Geospatial Institute, our role is to make that kind of progress possible: bringing together the right expertise, the right tools, and the right partnerships. This work reflects what’s possible when we combine new tools with a shared sense of purpose and remain focused on delivering advantage where it matters most.

The Taylor Geospatial Institute thanks Amazon Web Services for their partnership and the participating teams for their creativity and commitment. This report reflects the shared effort to advance generative AI for geospatial applications in service of the public good.
ABOUT THE challenge
In 2024, TGI and AWS launched the Generative AI for Geospatial Challenge — a first-of-its-kind national initiative to explore and expand “the art of the possible” in AI-powered geospatial problem-solving. Designed to catalyze experimentation, surface scalable ideas, and accelerate impact, the challenge opened the door to a diverse collection of participants and use cases. Their work offers a window into what generative AI can enable today — and what it may unlock tomorrow.

A total of twelve winning teams, drawn from TGI consortium institutions, commercial partners in the St Louis region, and experts from across the global community, were awarded a total of $1 million in AWS credits by a joint TGI-AWS review committee. Their projects applied generative AI techniques — ranging from large language models (LLMs) to diffusion models, multimodal tools, or novel architectures — to real-world geospatial challenges. Over four intensive months of work, teams designed, developed and demonstrated minimum viable products (MVPs). The result: a diverse portfolio of early-stage, deployable solutions that span sectors, technologies, and use cases.
Project Profiles
download Agenda
Cross-Modal Foundation Models for Real Time Decisions
Bedrock Research
Bedrock Research, Inc. - Matt Reisman, Kevin LaTourette
AgileView, Inc. -  Avi Lindenbaum, Peter Shagnea
Washington University in St. Louis - Nathan Jacobs
From Data to Insight: Predicting Tomorrow’s Landscapes
Clark Univeristy
Clark University - Hamed Alemohammad, Sam Khallaghi, Denys Godwin
Washington University in St. Louis - Nathan Jacobs, Srikumar Sastry, Dan Cher
Solving Missing Data Problems in Emissions Modeling
Crosswalk Labs
Multimodal AI for Rapid Disaster Response
Bringing Subsurface Intelligence to the Surface
Deep Earth
Deep Earth - Christie Capper, Albi Wiedersberg, Hamid Omidvar, Johannes Hansen
University of Illinois Urbana-Champaign - Yu-Feng Forrest Lin, Andrew Stumpf, Ashlynn Stillwell
Download the
Tracking Land, Food, and Health with AI
Earth Genome
Earth Genome - Ben Strong, Edward Boyda, Hutch Ingold, Mikel Maron
Download the
AI-Powered Mapping from Space: Custom 3D Maps on Demand
LuxCarta
LuxCarta - Sacha Lepretre
Download the
Cross-Modal Foundation Models for Real Time Decisions
University of Missouri-Columbia
University of Missouri Columbia - Dong Xu, Joe Soundararajan
Through Sensing LLC - Andrew Kalukin
Indiana University of Pennsylvania - Soundararajan Ezekiel
Download the
Automating Biodiversity Net Gain Assessments
Sparkgeo
Sparkgeo Consulting Inc. - James Banting
UK Multimodal AI Network
The Alan Turing Institute
University of Cambridge - Nataliya Tkachenko
Download the
Teaching AI to Understand Place and Context
Zephr.xyz
Zephr.xyz Inc. - Sean Gorman, Pramukta Rao, Kostas Stamitou, Joseph Strus, Ovidiu Voda
Download the
Real-Time Tools for Wildfire Damage Response
Avineon
Avineon, Inc. - Darryl Murdock, Oliver Morris
Midgard Raven - Adam Simmons
Analytic Folk - Chul Gwon
Bunting Labs - Brendan Ashworth, Michael Egan
Earth Fire Alliance
Heavy.ai
Download the
Property-Level Risk Insights from Predictive Modeling
Tera Analytics
Tera Analytics - Robert Carroll, Charles Mondello