March 2026
The single biggest barrier in GIS isn't the data — it's the query language. Spatial SQL is powerful but intimidating. Most team members who need answers from map data can't write ST_Within or ST_Buffer queries. They end up asking the one GIS analyst on the team, creating a bottleneck that slows everyone down.
Owl AI is our answer. It's an AI-powered assistant built into GISOwl that lets anyone ask questions about their geospatial data in plain English and get instant visual answers.
When you type a question into the Owl AI panel, the system analyzes your prompt alongside the metadata of your current map — layer names, column types, geometry types, spatial extent — and generates the appropriate spatial query.
Natural language to SQL: Ask "show me all parcels larger than 2 hectares within 1km of Main Street" and Owl AI generates the PostGIS query, runs it, and highlights the results on your map. You never see the SQL unless you want to.
AI Extensions: Build custom AI-powered widgets that live alongside your map. For example, an extension that automatically summarizes demographic data when you draw a selection area, or one that estimates flood risk based on elevation data in your layers.
AI Popups: Click any feature on the map and get AI-generated context. Instead of raw attribute tables, Owl AI synthesizes the data into readable summaries — "This parcel is zoned residential, 0.4 hectares, last assessed in 2024 at $320K, within a flood zone."
Prompting Tips: Owl AI understands spatial concepts natively. You can say things like "find the nearest hospital to each school" or "which neighborhoods have the highest tree canopy coverage" and it maps those to the correct spatial operations.
The immediate payoff is accessibility. Your entire team — project managers, field crews, executives — can explore map data without learning GIS software or SQL. They type a question and get an answer.
The long-term payoff is speed. Analyses that used to take a GIS specialist 30 minutes to set up now take 10 seconds. The specialist's time shifts from running routine queries to designing better data models and workflows.
We're expanding Owl AI to handle multi-step analysis chains — not just single queries, but workflows. "Find all properties within 500m of a transit stop, exclude those in flood zones, calculate average assessed value, and export the results." One prompt, multiple spatial operations, one result set.
The goal is simple: your maps should answer questions, not just display data.