Short Wave - Brain Implants Are Here — And Getting Better

Episode Date: September 10, 2025

Brain-computer interfaces (BCIs) are surgically implanted devices that link the brain to a computer. They can be helpful for people who’ve lost the ability to move or speak. And they’re making progress. New generations of BCIs could go as far as to detect a person’s inner monologue.But that progress is raising questions about the future privacy of our brains, and has some scientists asking, “What happens when you want to keep some things to yourself?”NPR brain correspondent Jon Hamilton talks to Short Wave’s Emily Kwong about the future of BCIs.Read more of Jon’s reporting on brain implants.Interested in more on the future of brain science? Email us your question at [e-mail] – we may feature it in an upcoming episode!Listen to every episode of Short Wave sponsor-free and support our work at NPR by signing up for Short Wave+ at plus.npr.org/shortwave.See pcm.adswizz.com for information about our collection and use of personal data for sponsorship and to manage your podcast sponsorship preferences.NPR Privacy Policy

You're listening to Shortwave from NPR. Hey, Shortwaivers, Emily Kwong here. If you're listening to this, you know this podcast has to travel through your ears to get to your brain. But why not take a more direct route, say from the Internet straight to your neurons? That's basically what we're talking about today. Brain computer interfaces with NPR's own neuromancer, John Hamilton. Hi, John. Hey, Emily.

Happy to be part of today's consensual hallucination, as William Gibson would put it. And with that, I guess we should probably stop with the dystopian sci-fi references. Let's keep them coming, John. I mean, Elon Musk is out here talking about using a brain computer interface, a BCI to connect us all to cyberspace, right? I mean, he has mentioned that. And his company, Neurrelink, has surgically implanted its brain computer interfaces, which we will call BCIs for short, in a whole bunch of people. Oh, and the Neurlink device, it's called telepathy. So I'm right. The future is here. You are certainly right about what one tech billionaire is saying,

but I am here to tell you that these interfaces are still a long way from putting our brain cells online. What they can do is pretty amazing. They can turn a person's thoughts into words or movement, and places like Neurrelink clearly hope they're going to do a lot more. So check out this promotional clip from Neurrelink. Imagine the joy of connecting with your loved ones, browsing the web, or even playing games using only your thoughts. This is made possible by placing a small,

cosmetically invisible implant in a part of your brain that plans movements. No, no, I'm not ready to enter the matrix, John. Not to worry, Emily. Right now, brain computer interfaces are still experimental. They're really only for people who are living with paralysis. Plus, Neurlink is really just one of the companies in this space, and some of their competitors are actually probably ahead. And the other places are much less focused on,

and, shall we say, joining the singularity. Well, today on the show, what brain computer interfaces can and cannot do. And why they're only a little bit scary, at least for now. You're listening to Shortwave, the science podcast from NPR. Okay, so, John, I'm really curious about what we're going to discuss today, brain computer interfaces. What does that mean? Yeah, okay. The term refers to technology that can detect and translate the electrical signals that are coming from,

your brain. Some of these devices can also send electrical signals to your brain. Wow. There are a few consumer products that are worn like a CAPRA headset and detect brain activity from outside the skull. We're not going to talk about those. So they're not that great. Some of them may help you meditate, but that's unclear. The devices we're going to be talking about are the ones that are implanted inside the skull, which is a much better place to listen to the brain's electrical activity. Some of these devices are put on the surface of the brain and other ones like the neuralink device, they actually thread electrodes into the brain itself. But why would someone want something inside their skull listening to their brain? So to answer that, I'm going to go all the way back to 2004.

One of the first people to ever get a BCI was a guy named Matt Nagel. He's a young man who got stabbed in the neck, sadly, and was left paralyzed. So he couldn't move his arms or legs. A team of scientists asked him if he wanted to try a device that was then called BrainGate. The idea was this might let him move a computer cursor using just his thoughts. I managed to find this old video of him testing that BCI with a researcher named Abe Kaplan. What are you going to do first? So keep in mind, this is all 20-some years ago. I'm going to open my email first.

Okay. The first one. You can open the first one. Was that the sound of the email opening? The ch-ching. Totally, yeah. He was able to do things that he had no ability to do with his body, but he could do with his brain. That sounds like life-changing technology. That's a pretty amazing. Okay, so this was in 2004. Has the technology advanced since then? In a lot of ways, yes. That braingate device that Matt Nagel used required actual wires to go through his skull.

Nowadays, devices are moving to wireless communication. They also use many more electrodes so they can listen. to the electrical activity of more neurons. That makes them more accurate. And the latest devices can do a lot more than just, you know, move a computer cursor. They can let a person who is paralyzed control a robotic arm accurately enough to sip from a cup of water. Whoa. Some can even provide that person with a sense of touch by sending signals from a robotic hand back to the brain. And BCIs can even decode speech signals in the brain, which means a person who, who can no longer talk because of a stroke or a disease like ALS, they can now communicate. That's, I've fully come around to this. This is amazing. So you're saying someone who can't control their own vocal tract can speak again with this assistance.

Pretty much. The BCI detects the words that a person wants to say and then it uses the computer to do the speaking. A lot of the work on speech BCIs has been done in California, including at the University of California Day. Davis. Let me play you a recording from UC Davis. This is of a man named Casey Harrell. He has ALS and can no longer physically produce words, but he can speak through a BCI that creates an artificial voice. I am. I mean, it's not super fast, but it's totally understandable. And this is coming from his thoughts? More or less, this particular BCI, it relies on signals from the most. motor cortex in the brain. So the part of the brain that usually tells the muscles that are involved in speech what to do. Oh, so the neurons use for signaling like your voice box, your throat, your mouth. That's what the BCI is tapping into. Exactly. So when someone attempts to speak, the brain starts sending those signals and the BCI is able to turn those signals into words.

This is something that several companies, including NeurLink, have included in their BCIs. But of course, it's all experimental. You know, we've had 20 plus years after the first BCI worked in a person, and there are still no FDA-approved brain computer interfaces that are on the market. I'm surprised by that. Why hasn't the FDA approved any BCIs? Well, they haven't really had any submitted. And I talked about the reason for that with Dr. Lee Hochberg. He's one of the scientists who've been in the BCI game since, you know, the beginning. Lee was one of the scientists who created the BrainGate device.

We heard Matt Nagel using a few minutes ago. I don't quite know how to summarize his resume, so I'm just going to play you what he said when I asked him to list all of his credentials. I'm a critical care neurologist and vascular neurologist at Massachusetts General Hospital. I'm also a professor of engineering at Brown University and a neurologist and researcher at the Providence VA Medical Center. In addition, I direct the Center for Neuro Technology and Neuro Recovery at Mass General, and I'm one of the co-organizers for the implantable brain. computer interface, collaborative community. This sounds like a Hollywood job, but it's real. It's a real job. You get the idea. He's got a lot of credentials.

Yeah. So what did Lee have to say about where we're at with BCIs, which sounds like not as far as we could be? One thing he said was he never thought BCIs would take so long to become a commercial product. But he also talked about some of the reasons why it has taken so long. It turns out it's really hard to take something that works in the lab, you know, with lots of computers and wires and technicians, and turn that into something that functions reliably for years inside a person's skull and can also communicate with their smartphone. I can't imagine why. Making this like portable, reliable forever sounds easy.

No, not at all. I mean, there are a few steps in there, right? And then there's the whole FDA approval process. Sure. It's slow, especially when you're dealing with technology like this that's completely new. And one more thing is that these devices are expensive. So companies that make them have to be thinking about whether they'll be covered by a person's health insurance. You know, despite all those things, Lee told me the industry is really finally close to getting actual products on the market.

Wow. I hope it's just in a few years that if I see and I'm taking care of somebody in the near ICU who suddenly lost the ability to move or lost the ability to speak, what I want to say is, I'm sorry this happened, but we have a technology that can restore your communication tomorrow. and really mean that if that conversation's happening on a Wednesday, that the technology will be deployed on Thursday and working. You're saying this could be the reality for some patients, like when? It could be just a few years. I mean, brain computer interfaces have been tested in dozens of patients now. There are at least half a dozen well-financed companies that are working on this technology.

Wow. And you've got artificial intelligence, which is making it much faster and easier to de-esuitable. code a person's brain signals. On the other hand, all that progress is raising some questions about what you might call brain privacy. This has been my lingering fear this whole time. Like, I want to be into this, but I'm worried about the intrusion of this technology. I mean, devices that read our thoughts and use that information, are we ready for that?

Well, that was exactly the question that some researchers had. They did a study that looked at this issue in BCIs that decode speech. So remember how those devices initially relied on a person trying to get their vocal apparatus to actually form a word. There's a new generation of devices that kind of goes the next step. It can actually detect a person's inner monologue. You know, that voice in your head that says things like, is this meeting ever going to be over? I don't have an inner monologue. Not everybody has an inner monologue.

This is something that it doesn't work for everybody. I'm safe for PCI. Yeah, exactly. But if you do have one, it's the easiest way to get sort of fluent speech out of a brain computer interface. Whoa. So what the authors found was there are ways to let people who are using a speech BCI switch off that inner monologue function, you know. But still, it's a little disturbing. And one of the people who's worried about this is Nita Farahani.

She's a professor of law and philosophy at Duke. She wrote a book called The Battle for Your Brain that talks about. about BCIs a lot. Nita told me she's all for better brain interfaces that will help people who are paralyzed. But she says things could get kind of weird when some of these functions come to consumer devices, like say a headset that lets you play a video game or post on social media using just your thoughts. The more we push this research forward, the more transparent our brains become.

And that's a double-edged sword. And we have to recognize that this era of brain transparency really is an entirely new frontier for us that we haven't even begun to grapple with. Yeah, I mean, brain transparency, brain privacy. This all sounds like where we started, dystopian sci-fi, except it's real. It totally does. And that is why Farahani and a lot of the scientists who actually work on BCIs, they're trying to ensure that, say, a marketing company can't just eavesdrop on our thoughts. Yeah, if I get ads delivered to my brain, I'm going to be mad. You and everybody

else, but don't think there isn't somebody thinking about it. I won't think anything. I'm going to keep this non-monologue on lock. Thank you so much, John Hamilton. It's great to have you on Shortwave. Always a pleasure, Emily. This episode was produced by Rachel Carlson. It was edited by Rebecca Ramirez and Burley McCoy, and it was fact-checked by Tyler Jones.

Quacey Lee was the audio engineer. Beth Donovan is our senior director, and Colin Campbell is our senior vice president of podcasting strategy. I'm Emily Kwong. Thank you for listening to Shortwave from NPR.