AI-designed paint can cool buildings during heatwaves
NEWS
Researchers have used artificial intelligence to develop a paint that can keep buildings cooler. The method, presented in Nature, may lead to both energy savings and more comfortable indoor climates. Max Yan at Umeå University has contributed to developing the method and interpreting the results.
The new paint, developed using AI, reflects solar radiation and helps buildings maintain a more comfortable indoor temperature.
ImageJohnér Bildbyrå AB
On a sunny day, buildings absorb much of the sun’s radiation and quickly heat up. Now, researchers from the USA, China, Singapore and Sweden have succeeded in creating a variety of advanced thin metamaterials which not only block solar radiation but also emit heat, hence lowering the indoor temperature. One of the materials can be sprayed or brushed from paint.
“A roof painted with the specially designed material maintained an indoor temperature 5.6 degrees lower than another building painted with conventional white paint,” says Max Yan, Associate Professor at Umeå University and co-author of the study published in the scientific journal Nature.
Machine learning creates complex structures
Metamaterials are artificially engineered materials with properties that are superior to those found in natural materials. Developing cooling metamaterials has previously been a time-consuming process where researchers had to rely on trial and error. The new method is based on machine learning and artificial intelligence, which in just a few days can generate thousands of complex structures with the desired properties of reflecting solar radiation and emitting thermal radiation. The best candidates are then identified in terms of mass producibility and cost.
Max Yan, Associate Professor at the Department of Applied Physics and Electronics.
ImageHans Karlsson
The new paint can provide significant energy savings by reducing the need for air conditioning. Calculations show that if the paint is applied to the roof of a typical four-storey apartment building with a roof area of 780 square metres in Bangkok, it could save about 12,000 kilowatt hours of electricity annually.
In addition to being used on buildings, the new paint can be applied to other objects, such as cars, trains, or machines. The need for effective cooling is increasing as the climate changes and heatwaves become more frequent and intense.
Saves energy in a warming world
“This type of material can contribute both to improved living environments and considerable energy saving,” says Max Yan.
In the longer term, the technology might be used for regulating the climate on a global scale.
“Will massive deployment of this type of radiative cooling metamaterial help to slow down global warming? I have not spent much time on this question, but I would say it is worth investigating,” says Max Yan.
In countries with cold winters, like Sweden, the AI approach can be adapted to design thin materials or paints that can help retain heat in buildings.
About the scientific article
Xiao, C., Liu, M., Yao, K. et al. Ultrabroadband and band-selective thermal meta-emitters by machine learning. Nature, 2 July 2025.