Roads Generated by Artificial Intelligence in OpenStreetMap
Keywords:
roads, artificial intelligence, OpenStreetMapAbstract
A concise summary of the article that analyzes the growing application of artificial intelligence (AI) in the generation of geospatial data and its input into OpenStreetMap (OSM). The problem of non-transparent import of AI-generated data without a clear declaration of the source is highlighted, which is against the principles of OSM. The cited research examines the possibility of recognizing AI-generated roads according to geometric characteristics. The key hypothesis is that algorithms create more uniform number of nodes than humans, whose work is more variable. The analysis is based on data from Malaysia and Tanzania, including AI-generated and traditionally mapped roads. Geometric methods for identifying AI-generated data have potential, but are still underdeveloped.
References
Andorful, F., Herfort, B., Melanda, R. A., Antonio, N. D., Zipf, A., Camboim, S. P. (2026): Transparency and Trust in Collaborative Mapping: Concerns and Dilemmas in AI-Assisted Road Integration within OpenStreetMap, Annals of the American Association of Geographers, 116, 3, 675–696. https://doi.org/10.1080/24694452.2025.2589286
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Nedjeljko Frančula

This work is licensed under a Creative Commons Attribution 4.0 International License.
