LO
GIS analyst / cartographer — tasks, one by one
The unit of analysis is the task, not the job title. Each one below carries its direction, whether the judgement rests on evidence or on platform inference, the reasoning, and what it does not establish.
Every task on this page#
Digitising features from imagery
Automating✓ Evidence-backedTracing buildings, roads, water and land cover from aerial and satellite images into map data.
At national scale, extraction is moving to models. France's national mapping agency says AI cut the production time of its large-scale land-cover map by a factor of three — one year per department instead of three — and halved the cost, and Microsoft publishes machine-made building footprints it says meet or exceed the quality of hand-drawn ones. At the level of one mapper it is less clear: in a controlled experiment with 26 participants, manual mapping was faster and more accurate overall than an AI-assisted editor, whose most common error was merging several buildings into one.
An agency's account of its own production, a vendor's claims about its own data, and one small experiment; none measures mapping jobs.
Spatial analysis
Being augmented≈ Platform inferenceRunning overlays, buffers, site selection and other analyses to answer a question.
AI agents can run the routine steps but struggle with the reasoning. A GIS agent built into desktop GIS showed a high success rate in tool selection and code generation for basic and intermediate tasks, while challenges remained for complex tasks. A benchmark of 50 geoprocessing tasks found the best model produced valid workflows 95% of the time, but spatial relationship detection and optimal site selection remained the most challenging, and its authors call for rigorous evaluation before claims about full GIS automation.
Benchmarks in test settings; valid workflows are not the same as correct answers, and neither measures analysts' work.
Spatial databases and map production
Being augmented≈ Platform inferenceMaintaining the spatial data and designing the maps and reports people use.
Demand for map information is rising. The US statistics bureau projects employment of cartographers and photogrammetrists to grow 7 percent from 2025 to 2035, much faster than average, without mentioning AI, and says increased demand for map information will require surveying and mapping technicians.
Projections for one country; no record measures AI in map design or database maintenance.
Quality checks and field validation
Still human-led≈ Platform inferenceChecking data against the ground and against standards before it is used.
The more machine-made data there is, the more checking there is to do. The US statistics bureau says that although drones and other advancements make some survey work more efficient, technicians will continue to be needed to review and interpret the output to ensure accuracy and completeness. Microsoft tells users never to import its building data without first checking the local quality, which varies between rural and urban areas.
A projection's reasoning and a vendor's advice to its users; neither measures how much checking is done.