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

TL;DR

This paper investigates how generative AI models, like ChatGPT, represent and reason about geography, highlighting the importance of understanding their world models beyond factual accuracy.

Contribution

The study introduces exploratory probes to analyze AI's geographic reasoning, addressing issues of default assumptions, brittleness, and deeper understanding beyond factual recall.

Findings

Models exhibit strong defaults and are sensitive to minor syntactic changes.

Distributional shifts can emerge from benign task compositions.

Factual recall does not necessarily imply genuine geographic understanding.

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?