The Very Real Case for Brain-Computer Implants

0:00Hey, this is Mike. Before we start, I want to take the chance to remind you that we want to hear from you. Do you have a question around AI or politics or privacy that has been on your mind or just a topic that you wish that we talked about on the show? If so, you can write to us at uncannyvalleyatwired.com. And if you listen to and enjoy our episodes, please rate the show and leave a review on your podcast app of choice. It really helps other people find us. Lauren, how are you doing? I'm hanging in there. Yeah. It's been a hectic week, but hectic in a good way, I think. Yeah, you've been very busy writing a lot of big scoops for Wired.com.

0:38Not a lot, but you know, enough. The business desk at Wired has been really busy this week. Of which you are a part. That's right. Yes. How are you doing? I'm doing great. Thanks. Wow, you rarely hear that these days. I know. We can probably just wrap the podcast now. How are you feeling about the fact that Katie is not with us this week? You know, I think that she is going to stun and surprise us all with our third weekly episode that she's planning. Right now we have two episodes a week of the show. She's working on this project that's going to bring a third episode to the show. And I think we're all going to be very impressed by it.

1:14So it is bittersweet. She's not here. But the sweet part is that we get a third drop in the feed every week. Yeah, I think it's going to be really good. Yeah. And also that means Katie still has time to run wired, which is another thing that she does on the side. I thought you were going to say run nine miles a day. That too. She also runs, runs, runs. But yeah, no, here we are, just you and I in the San Francisco office dealing with the fog. It's very foggy this time of year. People don't think that when they think of California, but it is. And is it just me and you? No, we have a guest. We have Emily Mullen on the show.

1:49You're kidding. No, we're going to bring her on in just a minute. Let's do it.

1:57This is Wired's Uncanny Valley, a show about the people, power, and influence of Silicon Valley. Today, we are talking about brain-computer interfaces. They're also known as brain-machine interfaces, or just BCIs for short. But whatever you call them, these are pretty incredible systems that allow direct communication between the brain and a digital device, like a computer or a phone. People who have had a BCI surgically implanted can use their thoughts as commands to make machines perform different tasks. There's currently a race underway in Silicon Valley to build a model that will stand out from the rest.

2:31And among the frontrunners are Elon Musk's Neuralink and a New York-based startup called Synchron. We'll dive into why the competition is heating up between these two companies and what the promises and limitations are behind this futuristic technology. I'm Marco Colori, Director of Consumer Tech and Culture here at Wired. I'm Lauren Good. I'm a senior correspondent at Wired.

2:57We are positively spoiled today to have a guest on the show who has reported on brain computer interfaces extensively, Wired's Emily Mullen. Hello. Emily, do you have a brain implant yet? No, I do not. Well, really, how committed are you to the bit then? I do not want a brain implant. No, thank you. Before we dive into BCIs, I would love to know what is the first thing that comes to mind when you think of brain-machine interactions, Lauren? I mean, for example, I think of RoboCop, the 1987 original by Paul Verhoeven, where it's just the RoboCop, it's just his head and his torso. And then like his limbs and all of his like running and walking are controlled by a computer that is implanted in his brain.

3:44I have never seen RoboCop. Oh, it's such a good movie. So cannot comment on that. What I think of, what I think of, well, this is a much headier response, but I think of all these promises that are being made around AI and healthcare and wondering if AI ends up being the sort of connective tissue between all of this that actually makes it viable. Like, I would never want one of these because it feels to me like a needs-based technology, not something you should just drill a hole into your brain in for fun. But if you get to the point where you need it, hopefully the technology is in place to actually help you, like, live aspects of your life that you wouldn't otherwise be able to live.

4:18Sure. Yeah. So I think we all have sort of a vague concept of what a brain-computer interface is and what it's for. Like a sci-fi concept of it. Yeah. But I want to ask you, Emily, is there anything remotely accurate in like the depictions we've seen in media? I mean, in the real world, how do these BCIs generally look and how do they work? Yeah, so I think the first thing to kind of clarify is that these are often called like mind reading devices. And I'm going to hopefully read your mind by clarifying this. And that, you know, a brain computer interface is not doing the kind of mind reading you might be thinking of.

4:56It's not extracting random thoughts from a person's brain. What it's actually doing is picking up on movement and tension in the brain. So when a person gets a BCI, they're asked to think about doing a specific action, such as opening and closing their fist. So a person who is paralyzed who gets a BCI may not be able to physically make that action, but there's still a signal that's firing up in the brain that's this distinct signal that the BCI is capturing. So the BCI learns to associate that pattern of brain activity with that specific action. And then the BCI is using AI to decode and interpret those neural signals.

5:46And then those brain signals get translated into a command. For instance, that opening and closing of a fist, that gesture, might correlate to a mouse click. Huh. Okay. So on the other end of that equation, there has to be some kind of physical or mechanical tool that then translates the movement. If you're a person, for example, who can no longer move your limbs and your brain isn't making that connection with the rest of your body, would you have to, say, for example, have like leg braces on that the computer would speak to in order for you to move? Right. Yeah. So Neuralink and actually other research teams before Neuralink have tried connecting robotic arms to BCIs to do exactly that.

6:29They've actually helped people feed themselves who were no longer able to do so because they were paralyzed. But this has happened, at least in the past, in these very controlled laboratory settings with these big clunky robotic arms. And that's not exactly practical for somebody to have at home. Okay, got it. So you can have a BCI system that isn't relying on physical hardware or robotics. What you're describing, the person is just sending signals directly to a computer and a computer is doing the task. Yeah, exactly. This technology is really meant to provide a direct link from the brain to an external device just to make that faster, that process faster and more lifelike and seamless for a person with a severe disability.

7:16So when we talk about BCIs, we hear about BCIs, the company that most people think of is Neuralink. And that's because they're a very high-profile company. They're run by Elon Musk. We all know who he is. But in your most recent reporting, you have focused heavily on Synchron. This is a startup that's been making waves lately because it has the support of several high-profile investors in Silicon Valley. and the fact that unlike other competitors, its technology requires relatively non-invasive surgery. The implant process is much simpler than Neuralink. Can you talk about the other ways that Synchron is different and why this company is such a big deal?

7:56Yeah, so of course, listeners have probably heard of Neuralink, but I guess to start off this conversation, we should go back to the OG BCI implant, which is the Utah Array. And the Utah Array has been like the mainstay of BCI research for the past 20 years or so. It looks like a little mini hairbrush, like the head of a hairbrush. It's got about 100 metal spikes protruding from it. And on the tips of those spikes are the electrodes which record the brain activity. And it is pushed into the brain tissue and records signals in that way. And that device has been implanted in a few dozen people over the years.

8:36but it's not wireless. It requires this really clunky setup that involves a pedestal atop a person's head. And beyond that, you need to get a craniotomy to get this chip implanted, and the chip can cause scar tissue to form over time, and that can interfere with the functioning of the device. So to be clear, a craniotomy is when you remove a piece of your skull with a bone saw temporarily to access the brain? Yes, that is correct. No fun. Oh, sounds slightly less pleasant than a colonoscopy. So all these reasons, you know, this BCI technology has been promising for a long time, but there are clear limitations as you can see.

9:19And this is why Synchron and Neuralink and others have been working to improve on that design. So I guess back to your original question, Mike, which what's the major difference with Synchron's approach? And the major difference is that it doesn't require brain surgery. That doesn't mean it doesn't require any surgery at all. It's less invasive. It's not completely non-invasive, but instead of going directly into the brain, Synchron's device is inserted into a blood vessel. It's a pretty unique approach. It actually is inserted into the jugular vein at the base of the neck, and then it's threaded up through the vein until it reaches the motor cortex.

10:06And that's the part of the brain that is responsible for movement. And this is a little device that looks like a heart stent used in cardiology. and the company actually, one of its founders is a cardiologist and that is what inspired this interesting design. It's dotted with electrodes, 16 of them that record the brain activity. During the same procedure, a second device is inserted into the chest pocket right below the collarbone and that's what processes the brain signals and then beams them out of the body via infrared. Wow. So wait, the chest pocket interprets the brain signals and then beams them out of the body, beams them using what, like Bluetooth?

10:54Infrared. Wow. It's like a remote control, like a TV control. Yes. And there's a receiver. There's like a second device that you just like hold up to your body to read the signals that are coming out of your brain. Yes. There's this paddle-like device that sits on the person's chest. And then that device, It's kind of a complex system. That device is connected via a wire to another device that is doing all of the processing and interpreting of those signals and translating them into commands. But right now, yes, it is a wired system. But in Synchron's 2.0 version, they are planning to remove that wire and have that communication happen via Bluetooth so that a person doesn't have to be, you know, physically sort of tethered to the system.

11:44And Emily, you wrote about in your story on wire.com how Synchron is also being built to work alongside other consumer technologies too. Can you talk about this? Yeah. So it seems like Synchron is really thinking about the ways in which it's eventual users will want to use technology in their everyday lives. So for instance, they have connected their BCI to Amazon Alexa. So their users now can use Alexa with just their thoughts instead of voice control. And you might be thinking like, well, why would you do that when you can just use your voice? But, you know, some ALS patients, some people who are paralyzed do not have the use of their voice.

12:24So that's where this would come in handy. Last year, Synchron also rolled out a feature to allow users to use an AI chatbot operated by OpenAI. And that's to, you know, again, facilitate communication. and Synchron also has connected its device to the Apple Vision Pro, which obviously if you are paralyzed or severely disabled, you might need somebody to put that device on for you. But the patient that I spoke with for my story on wire.com, he uses the Apple Vision Pro to kind of transport him to places around the world that he will never get a chance to travel to. And he actually really loves using that device.

13:09Wow. Apple Vision Pro. I take back everything I ever said about Apple Vision Pro. Let's talk a little bit more about a person who you profiled as part of your story. His name is Mark Jackson. He's a 65-year-old man from Pittsburgh, Pennsylvania, who is living with ALS, also known as Lou Gehrig's disease. It's a disease that causes the person suffering from it to gradually lose all of their muscle functions. He's one of 10 people that Synchron chose to outfit with its device in a clinical trial that started back in 2023. So he's using the device to talk to Amazon Alexa. He's using the device to control an Apple Vision Pro.

13:48How else is he using it? So he's also able to do things like play computer games. And when I visited him recently, he was playing this computer game that was sort of reminiscent of Pac-Man. He was in control of this little white circle and he had to drive it into a target and miss these blue obstacles. And he can also use the BCI to do things like write emails or shop online, text message. He can do his online banking. And as we mentioned, he can use the Apple Vision Pro. And Synchron is also in talks right now with other tech companies to integrate more types of consumer technology. So as we said at the top of the show, it's not reading his thoughts.

14:38Like when he's playing the video game, he's thinking of making a fist and releasing his fist and then making a fist again. And that's sending certain signals through the motor cortex, which is what the device is picking up. So I'm assuming that there's probably some sort of motor cortex actions he's doing to write an email or send a text message. Like he's not just like thinking of words and then it's writing them down. Yeah. So he's thinking about very specific commands that the BCI is picking up on. You know, yes, no, or selecting from certain letters, that kind of thing. It sounds like based on your reporting that Synchron is really focused on people who are living with severe disabilities.

15:21Elon Musk, on the other hand, with Neuralink, has talked about this vision of sort of a transhumanist future where the mind and machine meld. Not that that sounds dystopian in any way. So they do have slightly different focuses, is my understanding. Can you talk about that a little bit? Yeah, so we know Musk obviously has a tendency to present these very grandiose visions of his technology. Does he? Does he do that? Sometimes. But, you know, like any medical device company, Neuralink will need to prove a medical case to the FDA to get its product to market. So, yeah, there's what Elon Musk says.

16:01And then there's sort of the reality of getting a product to market. And I do think he sees his technology as something that will be adopted by everyone someday. I don't think Synchron sees its technology in that way. But I think like Synchron, Neuralink is first focusing on restoring digital autonomy to people with severe physical disabilities. And, you know, while Neuralink is obviously much more explicit about a future in which BCIs are more widely adopted, I don't think that's necessarily Synchron's goal. So Synchron has a major trial coming up in 2026, and it plans to enroll between 30 and 50 patients for the next round of testing.

16:50What happens next after they do this trial? How fast do you think that BCIs could become widely available to the public for anybody who needs one to get one? Yeah, so this kind of pivotal trial is really what Synchron needs to show the FDA that their device is safe, it's effective, and it can restore certain functions to people. In the early feasibility trial they did, they enrolled 10 total patients. There were four initial patients in Australia and six in the U.S. that got the Synchron BCI. And they followed those patients for a year after implantation. And they showed that there were no major safety events, which when you're talking about putting anything in the body, Of course, there can be risks like infection.

17:41As we talked about before with putting something in the brain, there's a risk of brain tissue damage. With the Synchron device, it's more about the possibility of blood clots or blocking a blood vessel. So that's what they want to look for in a bigger trial. And then the question in a bigger trial is sort of, well, how do you measure the effectiveness of a BCI? And the problem is there's never been a big pivotal trial of a device like this before, a device designed to restore certain functions. So one of the challenges right now is that these companies are working with the FDA. There was an FDA workshop back in the fall to get together researchers and BCI companies to talk about what are the clinical outcome measures that we are talking about here.

18:34We can measure how fast a BCI is, typing speed, or how accurately it could decode speech, for instance. But how does that correlate with a person's quality of life or their autonomy, their freedom? How do they feel when they're using it? So these are all questions that these devices are going to face as they head into bigger trials. And Neuralink is hopefully going to get there too. So I think it's still going to be several years before one of these devices is on the market because there are still some open questions. And also, of course, you know, these devices are going to be expensive. So what is the threshold of benefit where an insurer would cover this device?

19:21Emily, this is super fascinating. Your story is great. Everyone should go read Emily's story. Thank you for joining us in the Uncanny Valley. Come back with more goodness and weirdness soon. Thank you for having me. Thanks, Emily. We're going to take a break and we'll be right back.

19:44This week on the political scene from The New Yorker, Trump's rupture in the world order. Europe caught between two adversarial great powers. That's basically dialing back the clock to not only pre-World War II, but really it's a pre-20th century view of the world. And I would say it's a world of permanent insecurity that we're looking at. Join me, Evan Osnos, and my colleagues Jane Mayer and Susan Glasser every Friday on the political scene. Available wherever you get your podcasts.

20:23Welcome back to Uncanny Valley. Today, we are talking about brain-computer interfaces, also known as BCIs, and why Synchron and Neuralink are representative of two different approaches to the future of this technology. So, Lauren, we just had Emily on. She told us all about Synchron and the challenges and where the technology is and how Neuralink shapes up. Completely wild. What do you think? Mind blown. You know what I think? My first reaction upon hearing about this is this is the real biohacking. One of our earlier episodes of Uncanny Valley, which everyone should go back and listen to, was about the quest to live forever in Silicon Valley and all of the weird biohacking stuff that folks like Brian Johnson or even like the Jeff Bezos of the world are doing.

21:10And of course, there's been all this lure around like blood boys, right? Silicon Valley venture capitalists taking the blood of young interns to make themselves more youthful and vital. Of course, there's a lot of really important and significant investment happening in like cancer research and maternal health and things like that. But this is actually hacking the body to achieve outcomes that otherwise just would not be achieved. And it's high risk, it's invasive, it's expensive, it's all very experimental still. But it just seems like the upside, if you can get it right, is really significant for someone like the subject in Emily's story who is living with ALS and is paralyzed and is now able to experience the world through Apple Vision Pro.

21:56What? And, you know, iPhones and iPads and computers. And Alexa and AI. Yeah. And I think, you know, the idea of giving somebody back autonomy that they have lost, even if it's digital autonomy, is really important, right? Like, you know, you can communicate again. You can type using motor cortex commands, but you can type again. And giving somebody that ability to exist in the world is something that feels like a pretty significant step. Obviously, there are limitations. There is one anecdote in the story that Emily wrote about the subject, Mark, trying to do a Venmo payment using voice assistance technology.

22:42So, you know, like dictating to Siri, basically, like send a Venmo payment. But there is no way for the voice assistant to actually put in a reason for the payment. So there are places where like voice technology falls short and maybe, you know, BCIs are a way to sort of bridge that gap. But I do think it's novel that they're using iPhones and Apple Vision Pro as like the things that you can control. But really that speaks to how the company is thinking about how people are going to be using this in the future. And I think that's really smart on Synchron's part. And it seems like their approach to, even though it's still invasive, is less invasive.

23:19The idea of like threading something up through your jugular. Incredible. Yeah. It really is. Yeah. When the alternative is drilling a hole into your brain. And like Neuralink's interface is basically a mesh of little electrodes and they're robotically inserted into the brain tissue. And then they close you back up. And they put like a little sort of coin sized plug in your skull where the interface goes in. And that does not feel great. That feels like a stopgap. And going in through the vein and going up close to the brain gets you pretty close. It does not get you as close as Neuralink's BCI.

23:56So it seems as though like there's less that you can do with Synchron than you can do with a Neuralink implant. But it's still probably the option that most people would choose. It's fascinating. Now, I was reading that there's a pretty big difference in how much funding these two startups have raised. Synchron to date has raised just$145 million. Neuralink has raised$1.3 billion. And I have to wonder how much of that is just because of the halo of Elon Musk being associated with Neuralink. Or if there's something about their business that actually appears that much more viable to investors. I'm guessing it's the former.

24:37Yeah, it really feels like the halo around Elon Musk is undeniable. He can just drive, pun intended, all kinds of attention to the companies that he fronts. And I think part of that is just that he has been able to do so much in the industry through sheer will that people sort of believe that, hey, here's this crazy idea. If anybody can do it, then he can do it. Yeah, well, and he gets the engineers to do all the work for him, right? He motivates people. Yep. So how do you think AI chatbots are going to be used in brain-computer interfaces? This is the part that when you first hear about it, sounds like a little hokey.

25:12Like, oh, of course, ChatGPT is a part of this. But actually, it makes a lot of sense. When you think about the way that these language models are built to be predictive and actually anticipate what it is you are trying to say next, which we as humans who are able to vocalize and communicate the things that we want, you and I, like we even struggle with sometimes, right? When you think about that predictive technology being applied in a medical setting where someone actually cannot articulate what they need or what they want, it seems pretty smart. Yeah, and I think there are like specific things that we can point to besides just like the chat about use of AI.

25:50For example, there is this effect that BCI researchers refer to as the stadium effect. So if you think about how thoughts are presented in the brain, If you're inside a stadium, you can hear the conversations going on around you. Oh, right. Yes. This was an Emily story. This was really good. Yeah. You can hear the thoughts going on around you. Right. And you have pretty good knowledge of what's going on inside the stadium. If you're outside the stadium, all you really hear is the roar of the crowd. And you can probably discern when something big happens, like a goal is scored or something like that.

26:23So the Synchron device is a good example of something that is existing outside the stadium. Like it knows when the person who has it implanted is thinking about trying to make a fist and when they are thinking about releasing that. And it picks up these very like broad signals. So it's difficult to ascribe any kind of like specific intent beyond that. That's a place where AI can probably fill in the blanks. It can probably understand what the patterns are, what the person is trying to do. And if they keep repeating the same thing over and over again and not getting the outcome that they want, it possibly could be able to try different outcomes or somehow bridge that gap between what the person is really trying to do and what the signal should be.