ARM is getting into the brain game. The U.K. chip designer is one of the biggest players in the computing industry, especially when it comes to mobile devices. But it has now announced that it’s interested in building chips meant to go in between your ears, too.
ARM’s goals appear to be somewhat more modest, at least for now. The Softbank-owned company has announced that it will partner with the University of Washington’s Center for Sensorimotor Neural Engineering to help researchers there build chips that sit inside people’s skulls. The goal: to help people who are suffering from paralysis brought about by, say, stroke or spinal cord injury to move parts of their body again.
This story is only available to subscribers.
Don’t settle for half the story.
Get paywall-free access to technology news for the here and now.
To do that, they’ll together produce hardware that’s used for “decoding the complex signals formed within the brain, digitizing them so they can be processed and acted upon, with the end result of controlling the body’s muscle functions.” Over time, the organizations also hope to reverse the process so that the devices can process data to provide feedback directly to the brain, too.
Doing all that aboard a chip set that’s lodged inside your head raises some very particular challenges—because current implants farm those tasks out to an external computer, where constraints aren’t so tight. As ARM’s director of health-care technologies, Peter Ferguson, told the BBC: “The challenge is power consumption and the heat that generates.” What’s required, he says, is “something ultra-small, ultra-low power.”
ARM already has skin in that particular game. Its position as the go-to designer for the chips used in most of the world’s mobile devices has been cemented by the impressive energy efficiency of the chip sets it can create. In your smartphone that means the battery will last longer; in your head it means that things won’t get too hot.
Still, as others have shown, the processing of brain signals is just one computational challenge to solve. Recording neuronal activity in the first place requires a huge amount of computing power. In fact, it’s one of the focuses of the ambitious BRAIN initiative, which was launched by Barack Obama in 2013 to focus on building technologies that could interface effectively with neurons.
It’s currently unclear how far ARM’s ambitions might stretch beyond its stated aims, or whether it intends to get caught up in the techno-utopian visions of Musk and Zuckerberg. But either way, it’s setting itself a tall order. Still, it does at least appear to be tacitly aware of the magnitude of the task: it’s planning on working with the team at the University of Washington for the next 10 years.