This research introduces autonomous GIS, a novel paradigm where large language models (LLMs) serve as decision cores to independently generate and execute geoprocessing workflows.

• LLMs can represent and compute geographic information in innovative ways
• Autonomous GIS systems can perform spatial analysis with minimal human intervention
• This approach fundamentally transforms traditional geographic knowledge production
• Engineering implications include new system architectures that integrate AI with spatial capabilities

For engineering teams, this research opens possibilities for developing next-generation GIS platforms with embedded intelligence that can dramatically reduce manual workflow creation and enhance spatial analysis capabilities.

GIScience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS