Geospatial　Artificial　Intelligence　(GeoAI),　an　interdisciplinary　field　integrating　geographic　information　science　with　artificial　intelligence　(AI),　has　emerged　as　a　transformative　force　in　advancing　urban　computing　technologies　and　applications.　By　synergizing　the　inherent　spatiotemporal　characteristics　of　geospatial　data　with　AI＇s　advanced　inferential　capabilities,　GeoAI　provides　innovative　methodologies　for　addressing　multifaceted　urban　challenges.　Therefore,　we　first　systematically　examined　the　core　technological　components　of　GeoAI,　encompassing　geospatial　data　representation,　spatiotemporal　interpolation　and　prediction,　geo-related　knowledge　graphs　and　pretrained　spatiotemporal　foundation　models.　Then,　we　analysed　GeoAI＇s　implementation　in　urban　computing　through　four　representative　domains,　including　including　intelligent　transportation　systems,　environmental　surveillance,　public　safety　enhancement,　and　sustainable　urban　development.　Finally,　we　concluded　key　challenges　in　GeoAI-enabled　urban　computing,　emphasizing　the　integration　of　deep　learning　and　knowledge　graph,　interdisciplinary　collaboration　for　intelligent　solutions,　risk　mitigation　of　deceptive　spatiotemporal　data,　and　the　incorporation　of　human-centric　principles　in　GeoAI　technologies.