GIS Trend 1

From mapping places to understanding patterns

Geographic information systems (GIS) have long helped us organize, map, and analyze information tied to locations.

GIS trend from the history

Figure: Sikder et al. (2025), full paper

GIS is evolving beyond traditional desktop-based workflows and map-centric interfaces. The emergence of Large Language Models (LLMs) has introduced a new paradigm: Conversational GIS, where users can interact with spatial data, query geospatial services, and perform analyses using natural language.

But the next step is already taking shape: Agentic GIS.

While Conversational GIS focuses on understanding user instructions and translating them into spatial operations, Agentic GIS goes further. It enables AI agents to plan workflows, select appropriate tools, execute geoprocessing tasks, interpret results, and iteratively refine analyses with limited human intervention.

This transition marks a shift in how we interact with geospatial technologies:

Traditional GIS: Users operate tools and configure workflows.

Conversational GIS: Users describe tasks; AI translates them into GIS operations.

Agentic GIS: Users define objectives; AI agents plan, execute, evaluate, and coordinate workflows.

The underlying transformation is not simply about adding chat interfaces to GIS. It is about connecting LLMs with geospatial data infrastructures, APIs, processing engines, knowledge graphs, and interoperable standards such as OGC APIs.

One Thing: human oversight, reproducibility, data provenance, and transparent decision-making remain essential, particularly for complex spatial analyses.

The emerging question is no longer just how we can talk to GIS, but how GIS agents can autonomously reason and act on spatial data to support evidence-based decisions. We, GIS practioners and professions need to think critical but responsible how to shape the future.