Hardware That Learns: UCLA-Rooted Nanowire Networks Point Toward Physical AI

A new review article published in the journal Nature Reviews Physics lays out an emerging computing platform in which the hardware itself acts as the neural network, with no neural network software running on top of it. In this approach, computation occurs directly within the structure of a self-organizing material whose physical connections are measured on the scale of billionths of a meter. The review, titled “Self-organizing memristive networks as physical learning systems,” was published on 18 September 2026 and written by an international team of scientists who have developed and advanced the platform over the past decade and a half.

The work is rooted in research at the University of California, Los Angeles. Adam Stieg, a UCLA research scientist and associate director of the California NanoSystems Institute at UCLA, and CNSI member James Gimzewski, a distinguished professor of chemistry at the UCLA College, were lead investigators behind one of the first studies to introduce the approach. In recent years, Stieg has collaborated closely on further developing the technology with review co-author Zdenka Kuncic, a physicist at the University of Sydney in Australia. Theoretical physicist Francesco Caravelli of the University of Pisa in Italy is the first and corresponding author of the review, and other co-authors are experimental physicist Gianluca Milano of Italy’s National Institute of Metrological Research and physicist Carlo Ricciardi of the Polytechnic University of Turin in Italy.

“Silicon-based electronics have shaped how we think about computing, but they’re not the only way to do it,” Stieg said. “In our systems, the model evolves in the physical network itself. It adapts and changes.” That distinction captures the core idea of the platform: software and hardware are no longer separate. Instead of a program executing instructions on a passive chip, the material establishes and modifies its own physical connections in response to the signals it receives, so that learning and computing happen in the same physical substrate.

The motivation comes in part from the mounting costs of conventional artificial intelligence. Digital computing using silicon chips has transformed nearly every aspect of modern life and enabled the remarkable growth of AI, but as AI models scale up, that growth brings increasing demands for energy, water and computing infrastructure. At the same time, many emerging applications, from satellites and robots to distributed sensors, need to process information where it is generated, often with limited power and connectivity. The review authors position self-organizing networks as a complementary approach that could one day work alongside silicon-based digital computers to help mitigate these challenges, rather than as a replacement for them.

Over the last 15 years, researchers have uncovered rich collective behavior in these systems that can be harnessed to process complex information in real time with low power demands. The self-organizing networks are inspired by certain aspects of the human brain’s cortex, the seat of perception, cognition and reasoning. Like the brain, the systems process complex data without using much power. According to the review paper, self-organizing networks have performed various benchmarks for machine learning, such as speech and image recognition, in real time by exploiting the physical behavior of the network rather than running a conventional neural network model in software. The advantages, the authors argue, could help advance physical AI, meaning systems in which computation and learning are embedded directly into the physical hardware that senses and interacts with the world.

There are two distinct ways of designing these systems. UCLA-developed nanowire networks were first introduced in 2011, and nanoparticle networks were originally unveiled in 2013 by a team including review co-author Simon Brown, a physicist at the University of Canterbury in New Zealand. At a simplified level, the nanowires or nanoparticles can be compared to neurons, while the changing electrical connections among them resemble synapses. As in the brain, these systems process information in the form of electrical signals, and those signals lead to new connections forming. Frequent stimulation establishes persistent connections, analogous to memory, while a lack of stimulation degrades connections, analogous to forgetting.

“I’m less interested in what the system is made of than what it can do,” Stieg said. “Can it respond, adapt and process information in some of the ways biological systems do? If so, there may be many different materials that can get us there.” The researchers aspire to produce technology with brain-like capabilities to process changing streams of information efficiently while operating with far lower power demands than conventional AI hardware. At the same time, they are careful to emphasize that the brain metaphor has its limits. “The original inspiration was the brain, but we were never trying to build one,” Stieg said. “I don’t think of the brain as a computer, and I don’t think we can reproduce the full richness of a biological system. The goal is to understand which properties make biological systems so effective and see whether we can build those into engineered systems.”

A major potential application lies in edge computing, the processing that happens at all the spots where sensors gather data from the real world. Today’s sensors can generate enormous amounts of data, but often only a small fraction is ultimately useful. Conventional systems typically digitize those data and then rely on algorithms to recognize patterns, identify important features and extract the information needed for AI to make sense of it. That approach can be especially costly at the edge, where energy, computing power and communications bandwidth are limited. Satellites, for example, may collect far more data than they can efficiently transmit to Earth, requiring information to be compressed, filtered or reduced before transmission.

The review authors envision a future in which self-organizing physical networks become part of the computing process itself. Rather than simply serving as hardware on which software runs, these networks can adapt their structure in response to incoming signals, learning and computing physically. Such systems could ultimately enable AI to operate continuously and locally in resource-constrained environments. “Most AI treats the hardware as a passive platform for running software,” Stieg said. “We’re asking what becomes possible when the hardware itself is adaptive, when the material reorganizes in response to information and becomes part of the learning process. That points toward a very different kind of AI: one that is physical, energy-efficient and able to operate and adapt locally.”

The review article also notes that the computing platform could be the subject of study for physicists, material scientists, chemists, neuroscientists and computer scientists. “To really take advantage of this approach will require synergy across multiple disciplines,” Stieg said. “You can see that in the wide-ranging expertise of my collaborators and co-authors, each of whom brings unique domain expertise to the table.” The authors acknowledge support from the U.S. Department of Energy, New Zealand’s MacDiarmid Institute for Advanced Materials and Nanotechnology, New Zealand’s Marsden Fund, the European Union’s Next Generation EU program and the European Research Council. As a forward-looking review rather than a report of a single new experiment, the article consolidates fifteen years of accumulated findings and outlines where the field may go next, as researchers continue to test how far self-organizing materials can carry computation out of software and into hardware itself.

Subject of Research: Self-organizing memristive nanowire and nanoparticle networks as hardware-based physical learning systems for AI.

Article Title: Toward physical AI: When the hardware becomes the neural network

Article References: Toward physical AI: When the hardware becomes the neural network. Original publication (Bibliographic details unavailable.)

Image Credits: AI Generated

DOI: Not provided

Keywords: physical AI, nanowire networks, nanoparticle networks, self-organizing systems, memristive networks, neuromorphic computing, edge computing, UCLA, California NanoSystems Institute, Nature Reviews Physics, brain-inspired computing, energy-efficient computing