🤖 AI Summary
Sweden lacks a national registry of architectural heritage values, which hinders the formulation of effective building retrofit policies. This study addresses this gap by pioneering the integration of multimodal large language models (MLLMs) with street-view imagery to predict heritage values for over 150,000 buildings nationwide—encompassing approximately 5 million square meters of heated floor area—using zero-shot learning. Beyond delivering critical data to inform national retrofitting initiatives, the research systematically examines the ethical implications, transparency, and policy impacts of deploying vision-based large language models in public governance. The findings illuminate both the transformative potential and inherent risks of such AI applications, offering an innovative paradigm for AI-driven cultural heritage management.
📝 Abstract
During 2025 and 2026, the Energy Performance of Buildings Directive is being implemented in the European Union member states, requiring all member states to have National Building Renovation Plans. In Sweden, there is a lack of a national register of buildings with heritage values. This is seen as a barrier for the analyses underlying the development of Building Renovation Plans by the involved Swedish authorities. The purpose of this research was to assist Swedish authorities in assigning heritage values to building in the Swedish building stock. As part of the analyses, buildings in street view images from all over Sweden (N=154 710) have been analysed using multimodal Large Language Models (LLM) to assess aspects of heritage value. Zero-shot predictions by LLMs were used as a basis to for identifying buildings with potential heritage values for 5.0 million square meters of heated floor area for the Swedish Building Renovation Plan. In this paper, the results of the predictions and lessons learnt are presented and related to the development of Swedish Building Renovation Plan as part of governance. Potential risks for authorities using LLM-based data are addressed, with a focus on issues of transparency, error detection and sycophancy.