AI-designed paint could cool buildings by up to 20C and slash energy costs, a new international study shows.
Researchers from the University of Texas at Austin, Shanghai Jiao Tong University, the National University of Singapore, and Umeå University in Sweden used machine learning to create coatings that reflect sunlight and release heat better than conventional paints. The result: substantially cooler surfaces under direct midday sun.
This isn’t just for walls or roofs. The paints could keep cars, trains, and electrical equipment cooler, helping reduce the escalating need for energy-hungry cooling in a warming world. 
Applied to a four-story apartment block, this AI-generated paint could save about 15,800 kilowatt hours annually in hot climates like Rio or Bangkok. Scale that to 1,000 blocks, and you could cut power use equal to what 10,000 AC units consume in a year.
The study appeared in Nature, highlighting how machine learning is speeding up material innovation across sectors—from magnets for electric vehicles by British startup MatNex to AI tools by Microsoft for new solar and medical materials.
University of Texas professor Yuebing Zheng, co-lead on the research, framed this as a game-changer for material design.
“Our machine learning framework represents a significant leap forward in the design of thermal meta-emitters. By automating the process and expanding the design space, we can create materials with superior performance that were previously unimaginable.”
“Now, we follow the machine learning output, [its instructions for] the structure and what kind of materials we should use, and we can get it right without going through many, many design and fabrication testing cycles.”
The AI cuts down what once took a month to a matter of days, unlocking materials that trial and error never found.
Imperial College London chemistry lecturer Alex Ganose notes the field is booming.
“Things are moving very fast in this space. In the last year or so there have been so many startups trying to use generative AI for materials.”
“The process of designing a new material could require the calculation of millions of potential combinations. AI allows material scientists to push through previous restrictions in computational power. It also means the traditional process of creating a material and then testing its properties can be reversed, with scientists able to tell the AI what properties they want upfront.”
AI-designed heat-reflective coatings might become a powerful tool against soaring urban temperatures and rising electricity bills. The tech’s speed and scale mark a major step for sustainable materials science.
