Towards Transparency in Collaborative Mapping: Detecting Machine-Generated Roads in OpenStreetMap

The influx of machine-generated data, especially from Geospatial Artificial Intelligence (GeoAI), has grown significantly in less than a decade. AI-assisted mapping, a technological innovation leveraging open GeoAI data, integrates human validation to update crowdsourced databases, such as OpenStreetMap (OSM) is , creating a sense of security and quality assurance.  However, OSM contributors have expressed mixed feelings about the presence of AI content in the database, raising questions about their underlying fears and whether it is possible to differentiate between these data after integration. To explore this, first, we analyzed discussions among OSM contributors and identified concerns about data authenticity, quality, and the balance between human input and AI contributions.  Secondly, we  investigated the extent to which machine-generated roads can be detected within OSM. 

This data represent data used in our paper titled "Towards Transparency in Collaborative Mapping: Detecting Machine-Generated Roads in OpenStreetMap". It includes OSM roads data from GeoAI and OSM human constributors.