A tiny change at the boundary between two oxide layers may point to a less power-hungry future for artificial intelligence. Researchers led by the University of Cambridge have built a nanoelectronic device that behaves a bit like a brain synapse, storing and processing information in the same place instead of shuttling data back and forth as standard computer chips do. That matters because today’s AI hardware burns through vast amounts of electricity, and the demand is still rising. The team’s device is a memristor, a component designed to mimic how neurons and synapses adjust their connections. In this case, the memristor is made from a modified form of hafnium oxide containing strontium and titanium. The work appears in Science Advances. “Energy consumption is one of the key challenges in current AI hardware,” lead author Dr. Babak Bakhit of Cambridge’s Department of Materials Science and Metallurgy said in a statement. “To address that, you need devices with extremely low currents, excellent stability, outstanding uniformity across switching cycles and devices, and the ability to switch between many …