This study explores the potential of object recognition artificial intelligence algorithms applied to social media content as a tool to design sustainable and customized urban tourism experiences. Using Milan as a case study, the method addresses overtourism, urban cultural decentralization, and highlights the potential of peripherical neighborhoods. Despite the high concentration of tourist activity in central Milan, the study reveals rich opportunities in the suburbs, including local architecture, parks, and cultural heritage. It further emphasizes the need for long-term strategies over temporary event-based tourism, especially in view of the Milan 2026 Winter Olympics. Despite the few limitations of this approach the achieved outcomes of this study are manyfold. The paper advocates for a polycentric and inclusive urban vision, where digital tools and human-centered urban design and planning merge to foster resilient, equitable, and culturally vibrant cities.