Progress in AI is accelerating rapidly, but there are many frontiers that remain unexplored – and perhaps none larger than space. By space, I mean the space all around us – geography – and the new frontier is the integration of AI and the geographic information systems (GIS) that handle the exponentially increasing amount of geospatial data in the world.
The combination of AI and GIS is required to accurately observe, classify, and predict changes to environmental systems, from watersheds to agricultural land to forests. Without the ability to see and understand these changes, it is exceptionally difficult for researchers, conservationists and policymakers to take effective actions to protect and conserve our lands and waterways.
This gap between information and insight is precisely what we hope to close through our AI for Earth program. And it’s a problem we’ve been partnering on solving for some time with Esri.
Today, we’re expanding this partnership with the general availability of the Geo AI Data Science Virtual Machine (DSVM) with ArcGIS, now available here. Using the Geo AI DSVM, AI developers can harness the geospatial capabilities of ArcGIS Pro – Esri’s premier desktop GIS software – and Esri developers can access all of Microsoft’s AI investments, infrastructure, frameworks, processing and services on the same platform.
Already, we’re seeing how transformative this integrated approach can be. We’ve been using the Geo AI DSVM with ArcGIS Pro on one our most difficult AI for Earth lighthouse projects to date—a land cover mapping pilot project that allows for the conversion of aerial imagery into usable land cover maps, and the classification of land cover on the fly with great precision. This is a substantial change from the current process, which takes months if not years and is done only about every seven years in the United states – not great for precision conservation.
Greater availability of Geo AI DSVM can help achieve this kind of change at scale. With real-time information, policymakers can begin to answer questions like: Where is erosion impacting our forests, oceans and wetlands? Where are the threats of pollution to a region’s water supply? Where are farmers’ crops most likely to suffer from the effects of global climate change? And those answers can spark more effective interventions that improve the health of our land and water.
We are excited to be with Esri today to launch this offering to the general public. We know this tool can turbocharge AI applications with geo-spatial data and analytics and we are convinced that combining AI and GIS is critical for producing a better understanding how and where our planet is changing, so we can deploy precise and effective solutions.
Read more about the offering in this blog post by Joseph Sirosh, corporate vice president, AI and research, Microsoft, who presented it today on the Esri main stage.
Want to start using GeoAI Data Science Virtual Machine? Visit the Azure marketplace here.
Ready to take your GeoAI research to the next level? Apply for an AI for Earth grant here.