A breakthrough in magnetism could make AI chips use dramatically less energy

Brown University engineers have developed a method to create ultra-thin films of multiferroic materials that retain their electrical and magnetic properties. This breakthrough could lead to the development of computer chips that require significantly less energy, benefiting AI and high-performance computing.
Why it matters
Reducing the energy consumption of AI hardware is a major challenge for the tech industry, and this material science advancement offers a potential solution.
A promising new approach to building computer chips could dramatically reduce the energy needed for computing, including the rapidly growing demands of artificial intelligence. At the center of the breakthrough is a special class of materials known as multiferroics, which combine electrical and magnetic properties that can be controlled together using extremely small voltages.
These materials could make future electronics far more energy efficient than today's technology. However, a major obstacle has prevented their use in conventional computer chips. When multiferroics are made thin enough to fit inside these devices, they typically lose the electrical and magnetic behavior that makes them valuable. Scientists call this problem the "dead layer" effect.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in