In this talk I will introduce the development of city science and argue that amongst the great diversity of models that are now available to simulate different aspects of cities, AI has always been there is the background, enticing us to think about how we build better if not the best city. The focus I will introduce is not about the material artefacts of the city - intelligent buildings, autonomous cars - but about how we use intelligence to build better models about how we simulate and invent to the future city. Cities like most things in life are intrinsically unpredictable over anything but the very short term, and I will sketch the state of the art with a view to commenting on the cour knowledge about exiting cities and what this all holds for the post-pandemic future