Artificial intelligence, and specifically the subfields of computer vision and machine learning, has become a topic with great potential for predicting reuse patterns in the built environment. With sensors that collect visual data becoming more readily available, new opportunities are created to digitalise the built environment by applying technologies from these fields. Applications include exploring the design space, monitoring construction progress, and improving building performance during operation. Using these applications to increase circularity in the built environment requires information about in-use building products and their attributes (e.g. type, material, size, geometry, condition, etc.). This information is a starting point for many downstream circular processes and a core component of circular databases, which can enable designers, constructors, and facility managers to follow a circular paradigm. Many advancements have been made in academia and industry towards extracting such information from visual and other building data, e.g. for the downstream processes of predicting material reusability or automating the maintenance of building facades. This chapter presents efforts on this front and highlights the gaps in adopting and utilising these technologies for the circular built environment, including challenges in developing comprehensive systems for their deployment and in robustly evaluating them. It also discusses business and organisational considerations with respect to adoption, utilisation, and development of the technologies in the circular context.