Machine learning offers scientists a powerful new way to explore superconductors, materials that conduct electricity with zero resistance. The international team demonstrated that AI can quickly narrow down the nearly infinite number of possible material combinations and identify the most promising candidates. This approach could dramatically speed up the discovery of new superconductors, according to Aalto University professor Paivi Torma, who leads the SuperC consortium.
Superconductors can conduct electrical current without losing energy, but only when cooled to extremely low temperatures where quantum effects emerge. These remarkable materials are already used in technologies ranging from quantum computers and medical neuroimaging systems to nuclear fusion reactors and maglev trains.
Despite their great potential, superconductors remain extremely difficult to discover. There are virtually infinite combinations of chemical elements that can form new materials, but only a few turn out to be superconductors. Those that have already been identified generally require expensive cooling systems to bring them close to absolute zero before they can exhibit their unique properties.
Scientists around the world are searching for practical superconductors that can operate at room temperature.
“Superconducting materials that can operate at room temperature will forever change the way we consume energy,” Tolma explains. “If such materials can replace ordinary conductors in applications such as computers and data centers, it could reduce global energy consumption and significantly reduce the thermal footprint of the ICT sector.”
AI and quantum physics working together
The SuperC consortium was founded in 2023 by Professor Tolma and an international group of leading physicists who share the goal of using quantum physics to help tackle climate change. This is the first global collaboration dedicated to the discovery of new superconductors, with the ambitious goal of discovering room-temperature superconductors by 2033.
According to Törmä, the combination of quantum geometry and machine learning provides a strong foundation for that exploration. The team’s latest study shows that the properties of the newly identified superconductors YRu3B2 and LuRu3B2 are due to electrons forming flat bands within a kagome lattice, a geometric arrangement inspired by traditional Japanese basket-weaving patterns.
To identify these materials, the researchers first used machine learning to quickly screen the vast number of possible element combinations. A special algorithm selected the most promising candidates and analyzed them using detailed quantum calculations to determine whether they could become superconductors.
Once the predictions were theoretically confirmed, Rice University collaborators synthesized the material by chemically combining its building blocks to create new compounds. The Rice team, led by Professor Emilia Morosan, has experimentally verified that both materials are indeed superconductors.
Proof-of-concept studies have recently physical review study.
A faster path to new superconductors
It is extremely difficult to fully understand superconductivity from a quantum mechanical perspective, and the search for new superconducting materials is time-consuming and computationally intensive.
“Over the decades, researchers have recognized more than 7,000 superconductors, most by chance,” Tolma explains. “The process of identifying potential substances is so computationally intensive that researchers have actually only been able to theoretically predict the survivability of about 20 of them.”
Tolma points out that even if a material looks promising on paper, it may prove impractical because it is too difficult to synthesize or impossible to produce on a large scale. Traditionally, evaluating a huge number of potential materials required extensive computing resources. The SuperC team’s AI-driven approach changes that process by focusing detailed calculations only on the most likely candidates.
“Our method uses machine learning-based pre-screening followed by targeted calculations of promising candidates. This approach will significantly speed up the discovery of future superconductors. Using machine learning, we may be able to push the number of materials that can be processed into the billions,” Tolma says. “This brings us an important step closer to discovering room-temperature superconductors.”
Looking to the future
SuperC research will be presented in a paper from Aalto University. Designing for a cooler planet The exhibition will be held from September 1st to October 30th, 2026 in the Helsinki metropolitan area, Finland.
The SuperC consortium is funded by the Kavli Foundation, Klaus Tschira Stiftung, Kevin Wells, Jane and Aatos Erkko Foundation, Keele Foundation, Magnus Ehrnrootth Foundation, and Neste and Fortum Foundation.
