Geography According to ChatGPT -- How Generative AI Represents and Reasons about Geography

2026-03-19

AI Summary

This study investigates how generative AI represents and reasons about geographic knowledge, and evaluates its potential influence on public spatial cognition. Through three prompt-engineering–based probing experiments, complemented by qualitative case analyses and model behavior assessments, the work systematically uncovers implicit biases in large language models’ geographic reasoning—namely, strong default assumptions, high sensitivity to input phrasing, and distributional shifts and geographic biases revealed under compositional tasks. The research introduces a novel evaluation dimension that moves beyond factual accuracy, emphasizing the need to critically examine the plausibility and reasoning limitations of the “world models” constructed by AI systems.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Understanding how AI will represent and reason about geography should be a key concern for all of us, as the broader public increasingly interacts with spaces and places through these systems. Similarly, in line with the nature of foundation models, our own research often relies on pre-trained models. Hence, understanding what world AI systems construct is as important as evaluating their accuracy, including factual recall. To motivate the need for such studies, we provide three illustrative vignettes, i.e., exploratory probes, in the hope that they will spark lively discussions and follow-up work: (1) Do models form strong defaults, and how brittle are model outputs to minute syntactic variations? (2) Can distributional shifts resurface from the composition of individually benign tasks, e.g., when using AI systems to create personas? (3) Do we overlook deeper questions of understanding when solely focusing on the ability of systems to recall facts such as geographic principles?

Problem

• geographic representation
• generative AI
• spatial reasoning
• foundation models
• AI understanding

Innovation

• generative AI
• geographic reasoning
• foundation models
• distributional shift
• syntactic sensitivity