Urban air pollution is a growing concern due to its adverse effects on public health and the environment. While effective, traditional air quality monitoring systems are limited in coverage, data granularity, and timeliness. To address these limitations, AI-IoT-based solutions offer a transformative approach by integrating the Internet of Things (IoT) with Artificial Intelligence (AI) to enable real-time, large-scale monitoring and predictive air-quality analysis in urban areas. This paper explores the implementation of AI-IoT systems in which IoT sensors collect real-time data on key pollutants (e.g., PM2.5, NOx, CO2) from multiple locations across a city. The data is transmitted via wireless networks to cloud platforms, where AI algorithms analyze it to detect pollution patterns, forecast air quality trends, and identify pollution sources. Machine learning techniques are used to predict future air quality levels, while anomaly detection models alert authorities to sudden pollution spikes. Additionally, edge computing is integrated to process data locally, reducing latency and bandwidth consumption. The significant results show that AI-IoT systems provide more precise, timely, and actionable insights than conventional methods. Cities that have implemented these solutions report improved air quality management, enabling them to take preventive actions such as traffic regulation or public advisories. These results highlight the scalability and flexibility of AI-IoT systems in urban settings. The implications for the field are substantial, as this technology can transform how cities monitor air quality, making urban areas healthier and more sustainable through smarter environmental management.