In the US and many other industrialized countries, heat events account for more than all other natural hazards combined¹ and cities are getting hit the hardest. Urban areas are most susceptible to heat stress due to the high density of buildings, lack of green space and water bodies, and waste heat from industry and transportation. This phenomenon is known as the “Urban Heat Island” effect and means that cities can be much hotter than suburban and rural areas, which is bad news for health, mental wellness and energy use. With more than half of us living in cities² — a number that's only rising — and with climate change turning up the heat, we've got a growing problem on our hands.

This is where AI and data come in, playing a key role in determining where hotspots exist and are likely to develop, helping urban planners and municipalities design cities that are more livable and resilient.

As a participant in the 2025 EY Open Science AI & Data Challenge, you will have a chance to use AI for good, working with satellite datasets to develop machine learning (ML) models to forecast temperatures at microscales across a city. Your models can help make the case for more green infrastructure, leading-edge building practices and urban design that brings cooling relief to vulnerable communities.

To join the challenge, if you haven’t already done so, register here. If you have previously registered, you can start the challenge by accessing the appropriate image on the Active Challenges page. All challenge submissions are due by March 20, 2025.