Designing sustainable urban intelligence: a green AI framework and survey

The accelerating pace of urbanization has intensified demands on energy, water, transportation, and waste management systems, necessitating intelligent solutions that are both effective and environmentally sustainable. Artificial Intelligence (AI) has demonstrated significant potential in optimizing urban operations; however, the increasing computational complexity and energy consumption of AI models pose new sustainability challenges. This paper presents a comprehensive survey of Green AI methodologies and introduces the Green AI Urban Framework (GAUF), a structured conceptual framework for integrating environmentally responsible AI into sustainable urban systems. The survey examines recent advances in energy-efficient model design, lightweight and adaptive learning techniques, data-efficient training, and resource-aware deployment strategies. Major urban application domains, including electricity demand prediction, water resource optimization, waste management, and smart mobility, are analyzed to identify current capabilities, limitations, and research gaps. Building on these insights, GAUF integrates three core components: sustainability-aware model selection, cloud–edge deployment strategies, and energy-efficiency evaluation metrics across the AI lifecycle. The framework also incorporates implementation and governance considerations relevant to urban infrastructure, sustainability reporting, and municipal adoption. GAUF aims to support researchers, practitioners, and policymakers in designing AI-driven urban systems that balance predictive performance, scalability, and environmental responsibility. By consolidating existing research and outlining future research directions, this study contributes to the development of sustainable AI practices as a core component of next-generation smart and resilient cities.