Scientists have developed an AI-powered method that could dramatically speed up the search for room-temperature superconductors—materials capable of conducting electricity with zero resistance without the need for extreme cooling. Researchers combined machine learning with quantum physics to rapidly identify promising materials, leading to the discovery of two new superconductors, YRu₃B₂ and LuRu₃B₂. The breakthrough could help transform energy transmission, computing, and other advanced technologies.
Traditionally, discovering superconductors has been an extremely slow process because researchers must evaluate an enormous number of possible material combinations using computationally intensive quantum calculations. The new AI approach first screens billions of potential candidates to identify the most promising ones, allowing scientists to focus detailed analysis only on materials with the highest likelihood of exhibiting superconductivity.
After the AI identified the best candidates, researchers confirmed their superconducting properties through quantum calculations and laboratory experiments. The newly discovered materials belong to a class known as kagome superconductors, where electrons move through a distinctive lattice structure that gives rise to their superconducting behavior. This successful demonstration shows that AI can significantly accelerate the discovery of novel quantum materials.
The research represents an important step toward the long-term goal of finding a practical room-temperature superconductor. Such a material could revolutionize electricity transmission by eliminating energy losses, reduce power consumption in data centers and computers, and enable major advances in technologies such as quantum computing, fusion energy, and magnetic levitation transportation. By combining machine learning with quantum physics, scientists believe they can dramatically shorten the time needed to discover the next generation of superconducting materials.
