The researchers then asked residents to rate how realistic and recognizable the images were.
Kim said these gaps in representation will become important as generative AI tools become more common in travel planning, urban design, marketing and public communications.
“People are increasingly relying on AI-generated content to learn about places,” said Kim, assistant professor in the Department of Geography in the School of Natural Resources and Environment and director of the Smart Cities for Good research group. “If smaller cities are not well represented in the data used to train these systems, the images people see may not reflect the true identities of those communities.”
The researchers surveyed 129 participants, asking them to rate the AI-generated images based on how realistic they looked and how well they captured each city’s identity. The images focused on elements such as landmarks, neighborhoods, trails, and waterfronts, using urban design principles developed by planner Kevin Lynch.
The study found that the AI struggled the most with landmarks and culturally significant features. In Blacksburg, for example, the university buildings were not adorned with Hokey stones.
Researchers also found that long-term residents were more critical of AI-generated images than new residents. Kim said this suggests that people with more local knowledge are more likely to notice inaccuracies or missing details.
