Cities around the world are facing increasing threats from extreme heat and poor air quality, both of which disproportionately impact vulnerable populations. In Manchester, a team of researchers is leveraging artificial intelligence and open data to develop innovative tools that help urban areas monitor, predict, and respond to these environmental challenges more effectively. By combining real-time environmental data—such as temperature readings, air quality measurements, and weather forecasts—with machine learning models, the team has created dynamic risk assessment platforms. These tools can identify hotspots within cities where heat stress and pollution levels are highest, often pinpointing areas with older infrastructure, limited green space, or dense populations of elderly or low-income residents. The AI systems are trained to analyze patterns across diverse datasets, including satellite imagery, traffic flow, building density, and historical health records. This allows them to forecast when and where extreme conditions are likely to occur, enabling city officials to issue early warnings, activate cooling centers, and adjust traffic management to reduce emissions during peak pollution events. A key strength of the project is its reliance on open data, which ensures transparency and allows other cities to adapt the tools to their own needs. By sharing methodologies and datasets, the Manchester researchers are promoting a collaborative approach to urban climate resilience. The initiative is already being piloted in several UK cities, with early results showing improved response times and more targeted interventions. As climate change intensifies, such AI-powered solutions offer a scalable way for municipalities to protect public health, reduce environmental inequities, and build more sustainable, adaptive urban environments.