AI is expanding rapidly, demanding ever more energy, but his team at the University of Twente is looking in a different direction. Instead of building ever bigger and faster servers, they are developing 'neuromorphic chips'. These are smart devices that do more with less energy, offering a radically new approach to one of AI’s biggest challenges.
"One AI prompt can use as much energy as charging your phone for ten minutes", Wilfred explains in the podcast. "The data centre is out of sight, but the energy bill is not."
Smarter chips that act more like neurons
At the NanoElectronics group in Twente, researchers are building analogue, trainable chips that behave more like networks of neurons than traditional processors. "Their behaviour is not fixed at design time", Wilfred explains. "You can tune them after production, just like you would train a neural network. By leveraging reconfigurable nonlinear processing units, we can perform complex operations more efficiently, reducing both the number of steps and the need to fetch data from memory repeatedly. The physics does the computing for you."
One application for this approach is speech recognition. Wilfred’s team recently filed a patent for a system that processes raw audio directly on their neuromorphic chip. This enables the performance of high-quality speech recognition with significantly smaller and simpler digital models. "We skip a lot of the heavy signal processing up front. No need to convert to digital or to the frequency domain", he says. "That saves a lot of energy and still gives good results."
These chips don’t just process information, they actually learn!
What happens when hardware starts to learn?
The idea of a chip that can adapt to its environment sounds powerful, and it is. But it also raises important ethical questions. "We are developing physical systems that can adapt and learn, essentially bringing artificial intelligence into the hardware itself", Wilfred notes. "If these systems become more autonomous, we need to ask hard questions about control, responsibility and unintended consequences." To address these challenges, his group works closely with ethicists and philosophers.
From data centres to real-world devices
Wilfred does not see neuromorphic chips as a replacement for digital computing. "We will always need precise, robust systems for a lot of tasks", he says.
Instead, he envisions adaptive chips as a complement, especially for data-intensive jobs where energy use matters. Think of edge devices in phones, cars, or even medical implants. "They are lightweight, fast, and do not rely on a constant internet connection. That is what makes them ideal for real-world environments", he adds.
Shaping the future responsibly
As AI hardware evolves, so do its societal impacts. Wilfred hopes more researchers will take this seriously from the start. "We do not want to build a powerful technology and only later ask whether we should have done it differently", he says. "That conversation has to happen now, as the field is still taking shape."
