Urban digital twins are virtual city replicas that can greatly support urban planning by simulating infrastructure and mobility scenarios. However, keeping a digital twin up-to-date with fine-grained, real-world urban conditions is challenging. This paper proposes a novel system that leverages multi-modal AI models to bridge the gap between physical urban data collection and a 3D city digital twin. In proposed approach, ordinary smartphones carried in vehicles act as mobile sensors, continuously capturing multimodal data (road images, GPS coordinates, and speed). Advanced vision-language models then analyze the data to automatically extract information from the traffic and road infrastructure and detect road anomalies. The extracted information such as the locations of traffic signs, traffic signals, road surface, and potential blind spots at intersections is geo-tagged and streamed into vision-language models to interpret data and stream human readable insights into the digital twin model. The case study is the digital twin of the city of Toronto. By aggregating data from many drivers and analyzing it (in post-processing for high accuracy), the digital twin evolves into a living model of the built environment. This enriched and dynamic twin provides urban planners with up-to-date insights on traffic signage, road conditions, and other relevant road infrastructure elements, enabling proactive maintenance and informed decision-making for city planning.