AI-Driven Adaptive Urban Lighting for Reducing Light Pollution and Energy Consumption in a Multi-Level Perspective

Urban lighting systems contribute significantly to energy consumption and light pollution, raising environmental and societal concerns. This paper explores the potential of Artificial Intelligence (abbreviation: AI)-driven adaptive urban lighting as a sustainable solution, framed within a multi-level perspective on socio-technical transitions. At the landscape level, increasing urbanization and global sustainability targets exert pressure for energy-efficient practices, while traditional street lighting regimes remain largely rigid and resource-intensive. At the niche level, we propose a novel adaptive lighting system integrating real-time Internet of Things (abbreviation: IoT) sensor data and machine learning algorithms to dynamically adjust illumination based on traffic, pedestrian activity, weather conditions, and ambient light. Studies demonstrate that the proposed approach can significantly reduce energy use while minimizing light pollution, without compromising safety or visibility. The results indicate that such niche innovations, supported by AI and renewable energy integration, have the potential to influence broader regime change and contribute to sustainable urban development. This research highlights the importance of combining technological innovation with socio-technical frameworks to address pressing urban environmental challenges, offering insights for policymakers, urban planners, and energy managers seeking to balance efficiency, safety, and ecological impact.