Quantum materials could cut AI data centre energy use

Driven by a passion sparked by pop-science and YouTube videos, PhD student Jean-Félix Milette is researching quantum materials that could dramatically reduce the energy demands of artificial intelligence data centres. Milette, the first in his family to attend university, focuses on topological insulators, materials that insulate internally but conduct electricity on their surfaces with unusual resistance to imperfections. “The goal is to continue scaling devices down while maintaining or even increasing performance and efficiency,” Milette explains, envisioning a future where less heat generation translates to reduced energy consumption and freshwater use in AI systems. In fall 2026, Milette will further his work at Oak Ridge National Laboratory through a competitive research internship.

Topological Insulators Enable Efficient Quantum Memory Topological insulators present a unique pathway to enhanced magnetic random-access memory (MRAM), potentially overcoming limitations in conventional RAM designs. These quantum materials insulate internally but allow electrical current to flow unimpeded along their surfaces, a characteristic stemming from their distinct electronic structure; this surface conductivity remains remarkably stable even with material imperfections that typically impede electron flow. Jean-Félix Milette and colleagues are combining these materials with magnetic layers to create a prototype MRAM device, aiming to reduce energy consumption in data storage. The unusual resistance to imperfections is critical because smaller devices require increasingly precise manufacturing, and topological insulators offer a degree of robustness against these challenges. By minimizing heat generation within memory devices, the need for extensive cooling can be lessened, conserving both energy and a vital natural resource.

The team’s work builds on the principle that reducing heat output at the component level offers a more sustainable solution than simply improving cooling infrastructure.

Oak Ridge National Laboratory deployed Pathfinder, a 20-qubit IQM Radiance system, on July 8th, July 13th, and July 24th, 2026, demonstrating a commitment to integrating quantum computing with high-performance computing facilities, providing a platform for testing and refining these new materials. This placement, supported by both the Mitacs Globalink Research Award and the Graduate Research at Oak Ridge National Laboratory program, will allow detailed analysis of the quantum magnetic interfaces within the MRAM prototypes. Oak Ridge installed a real-time quantum error correction system, the Riverlane Deltaflow 2, by the end of September 2025, underscoring its position as a leading hub for quantum technology development. “Part of what I’ll be doing is bringing knowledge back to Canada and helping expand the research community here,” Jean-Félix stated, emphasizing the broader impact of his work and the importance of international collaboration. He credits Ryan Plumadore with a key moment in his career, recalling, “He told me I had great potential for research,” and adding, “That was probably the point when I started thinking that maybe I could actually do it.” The potential benefits of this technology are significant, offering a pathway towards more sustainable and efficient computing infrastructure, particularly for the rapidly growing field of artificial intelligence. Researchers at Oak Ridge National Laboratory demonstrated a prototype system capable of live quantum network data and alert generation on July 11, 2026, showcasing the lab’s advancements in quantum networking and data processing. This work, combined with the development of quieter isotopes for quantum device fabrication on July 24, 2026, achieving a hundredfold reduction in noise, positions Oak Ridge as a key player in the advancement of quantum technologies and their integration with existing computing systems. The goal is to continue scaling devices down while maintaining or even increasing performance and efficiency. Jean-Félix Milette, PhD Student Magnetic RAM Reduces AI Data Centre Cooling Needs Magnetic RAM, using the unique properties of topological insulators, presents a pathway to significantly lessen the thermal load in artificial intelligence data centres. This configuration aims to create magnetic RAM, a potential successor to conventional random-access memory, and address the escalating energy demands of AI computation. The efficiency gains stem from the materials’ inherent characteristics; topological insulators minimize resistance, reducing heat generation within the memory itself. This commitment to quantum research is bolstered by partnerships with companies like NuScale Power, IQM Quantum Computers, and Riverlane, facilitating collaborative advancements in the field. Currently, AI data centres require substantial cooling systems, often relying on fresh water to dissipate heat generated by processors and memory. The development of low-heat magnetic RAM could lessen this dependence, conserving water resources and reducing overall energy consumption.

Oak Ridge National Laboratory’s recent advancements in isotope purification, achieved on July 24, 2026, producing silane and germane isotopes over 100 times more depleted of noise-inducing isotopes than commercial supplies, further support the creation of more stable and efficient quantum devices. This placement is particularly significant for Canada, which currently has limited neutron research infrastructure. That was probably the point when I started thinking that maybe I could actually do it. Source: https://www.uottawa.ca/en/news-all/could-quantum-materials-make-ai-more-energy-efficient More like thisDeep TechBristol and NQCC launch intensive quantum computing coursesDeep TechHands-on quantum learning coming to schools with new EPFL projectArtificial IntelligencePasqal and LG CNS build quantum-ready AI data centersQuantum Research NewsQuantum leaps in Canadian facilities enable nanoscale device buildingStay currentSee today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals. Tags: The Neuron With a keen intuition for emerging technologies, The Neuron brings over 5 years of deep expertise to the AI conversation. Coming from roots in software engineering, they've witnessed firsthand the transformation from traditional computing paradigms to today's ML-powered landscape. Their hands-on experience implementing neural networks and deep learning systems for Fortune 500 companies has provided unique insights that few tech writers possess. From developing recommendation engines that drive billions in revenue to optimizing computer vision systems for manufacturing giants, The Neuron doesn't just write about machine learning—they've shaped its real-world applications across industries. Having built real systems that are used across the globe by millions of users, that deep technological bases helps me write about the technologies of the future and current. Whether that is AI or Quantum Computing.