Geography According to ChatGPT -- How Generative AI Represents and Reasons about Geography
This addresses concerns for the public and researchers about AI's geographic reasoning, though it is incremental as it provides exploratory probes rather than new solutions.
The paper investigates how generative AI models represent and reason about geography, highlighting issues like model defaults, brittleness to input variations, and overlooked deeper understanding beyond factual recall.
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?