As part of King’s College London’s Safe, Trusted & Responsible AI (STaR-AI) Doctoral Programme, we are advertising a PhD position for a project entitled Explainable vision-language AI for trustworthy urban placemaking in collaboration with consulting company ERA-co [UK Students Only].
For details about the available funding and the application process, see https://www.findaphd.com/phds/project/explainable-vision-language-ai-for-trustworthy-urban-placemaking/?p193649
Application deadline:Â 2 March 2026
Background and context: ​The current wave of generative AI is increasingly discussed for its potential to address the complex challenges faced by cities (Luusua et al., 2023). Current vision-language models (VLM) and social media data enable the analysis of massive amounts of street-level visual and textual urban data for the characterisation of places, including the fine-grained detection of business and leisure activities, changes in the built environment, and deprivation (Wu et al., 2025). Despite their potential, existing AI solutions in urban planning continue to suffer from critical limitations: limited transparency, weak cultural sensitivity, and low public trust, particularly among vulnerable communities, as frequently observed in planning consultations (Lartey & Law, 2025).
Characterising intangible aspects like place character and imageability is a common concern in urban design theory. Yet, there is currently no robust method for integrating multimodal visual, textual, and geospatial data to support culturally grounded, transparent, and trustworthy AI for urban consultation and decision-making (Ye et al., 2025). In collaboration with consulting company ERA-co, this project addresses this gap by developing a framework for AI-driven place characterisation using VLMs, with the aim of strengthening trust in AI-supported placemaking.
Cultural Geo-Analytics Lab 