Urban energy and traffic systems are two highly interdependent components of smart city infrastructures, both of which operate under significant uncertainty caused by fluctuating demand, human mobility patterns, weather variability, and policy constraints. While Artificial Intelligence (AI) techniques particularly machine learning and deep learning have demonstrated strong predictive capabilities in these domains, their black box nature limits interpretability, trust, and adoption in real world urban governance. Methods: This study proposes an adaptive fuzzy hybrid artificial intelligence framework that integrates fuzzy inference systems with ensemble machine learning models to support uncertainty aware and explainable decision making in urban energy and traffic management. The proposed framework is validated using real world secondary data obtained from open government and smart city data portals, including urban energy demand, traffic flow, and environmental indicators. The primary objective of this research is to develop a robust and interpretable decision-support model capable of dynamically adapting to uncertain urban conditions while maintaining high predictive performance. Experimental evaluations demonstrate that the proposed fuzzy hybrid AI framework consistently outperforms standalone machine learning approaches in terms of decision stability, robustness under uncertainty, and interpretability across multiple urban scenarios. Conclusion: The findings indicate that adaptive fuzzy hybrid AI offers a practical, scalable, and policy aligned solution for urban energy traffic decision support, contributing to sustainable smart city governance and supporting evidence-based decision making in line with global sustainability agendas.