GIS managers have always needed unique skills to succeed in their role. But agentic AI in GIS comes with integration challenges, budgeting considerations, and team-building demands that didn’t exist even two years ago. To run and grow a GIS program in this new era, every manager must have and hone five new skills.
1. Design GIS infrastructure for agent use
Agentic AI introduces a major design challenge for GIS infrastructure: Your system must accommodate agents in addition to people—and AI lacks the context to navigate the peculiarities of geospatial data and tools.
This architectural shift has practical implications across several areas:
- Data readiness—Not all datasets are ready for an agent. Some have quality issues, while others have informal caveats or context that only your team understands. You need a way to evaluate which datasets are agent-ready and provide instructions for appropriate use.
- Metadata and documentation—There’s a difference between metadata written for a GIS analyst who understands conventions and metadata written so an AI model can decide whether a layer is appropriate for a given query. You must describe data and services to steer agent behavior rather than just inform human users.
- Tool and service provisioning—When an agent connects to your GIS environment, which geoprocessing tools does it have access to? Which data layers? Which services? This is like provisioning a new employee’s toolbox, except that a person will ask for clarification; an agent will use whatever you give it.
- Connectivity—GIS managers must understand how agents connect to and interact with spatial infrastructure. That includes API exposure and the protocols through which large language models (LLMs) call spatial services. You don’t need to build this yourself, but you need to know enough to make informed decisions. An agent that queries a published feature service, for instance, is different from one that has direct access to your geodatabase or geoprocessing tools. The scope of what it can read, write, and execute depends on how it’s connected. If you don’t understand those distinctions, you could inadvertently introduce usage and data access risk.
2. Decide what agents do and what people do
Five years ago, no GIS manager had to ask this: Which workflows stay human, which go to agents, and which are hybrids with human checkpoints along the way? Now, it’s a daily conundrum!
Currently, LLMs and the agents they power are weak at performing certain spatial tasks. When it comes to precise coordinate math, topological reasoning, and map projections, agents confidently produce wrong answers if you let them. Knowing where agents are likely to fail is a new form of technical literacy.
The division of labor between humans and agents is an operational design question with real consequences. If you get it wrong in one direction, you’ve automated workflows that need human judgment—and potentially introduced a cascade of errors. Get it wrong in the other direction, and you risk inserting so many human checkpoints that you recreate the manual process with extra steps.
For each workflow, assess the cost of an error. How well can an agent handle the spatial reasoning involved? Where should a human review before the output goes further?
3. Consider the economics of agent-mediated work
Traditional GIS expenses are mostly predictable. They are based on licenses, full-time staff, and cloud-based consumption costs. But AI agents take consumption-based outlays to another level. Every query, analysis, and output carries a cost tied to compute cycles, API calls, and token-based AI services. Those costs can scale in hard-to-predict ways.
This creates three economic challenges that GIS managers must get comfortable with:
- Take economics into account—Find out what a single agent-driven analysis costs, and compare that to a human doing the same work.
- Manage demand—When spatial analysis takes two minutes instead of two hours, demand surges. Managing that is a new operational challenge.
- Build the return-on-investment case—What is it worth when an agent that costs $1,500 per month frees up 15 hours of analyst time for strategic work?
4. Assure quality at agent scale
Because AI agents operate at unprecedented speeds, quality assurance (QA) can outpace your team’s capacity. How do you avoid this without rubber-stamping tasks just to get them off your desk?
When a GIS team produces five maps or datasets each week, it’s easy to keep up. But when an agent generates 500 outputs in the same period, QA becomes harder to handle.
GIS managers must design QA processes that work at agent speed and volume. One method is to shift from reviewing every output to defining error rates for different use cases, then randomly sampling outputs. Different use cases will naturally have distinct acceptable error rates.
This also means building audit trails. When an agent produces a spatial output, you need to be able to reconstruct what happened. What data did it consume? What operations did it perform? What data was the model trained on?
Previously, when something went wrong with a map or analysis, you could walk over to the analyst’s desk and ask what happened. In an agent-driven workflow, observability and traceability should be built into the process. Otherwise, your agent-driven spatial work might not be trustworthy enough to use.
without requiring a human to review every result.
5. Govern agents and mitigate risk
With agentic AI, agents themselves must be governed alongside geospatial data. This means dealing with new risks.
Who in your organization can deploy an agent that consumes your spatial data and services? What approval process do they go through? What are agents permitted to do—perform read-only queries or publish outputs, modify data, and trigger workflows? These questions need clear answers. Yet in most organizations today, nobody has written the policies.
As a GIS manager, you’re the best person to raise these questions because you understand what can go wrong with spatial data. While you’ll need stakeholders to weigh in, you should kick-start the conversation around governance for geospatial agents.
GIS managers must still understand GIS capabilities, articulate the technology’s value, manage costs, know the latest integrations, build teams, govern data, and track the direction of the industry. Those skills haven’t changed. But agentic AI adds new complexities that demand focus. If you build these skills now, you won’t just keep up with change—you’ll lead the way.
