WEBVTT - The Brain Implant That Could Change Medicine 0:00:15.356 --> 0:00:23.076 Pushkin. My guest today is a brain surgeon who also 0:00:23.156 --> 0:00:27.796 has a PhD in electrical engineering from MIT, which is 0:00:27.836 --> 0:00:32.076 to say he is extremely well prepared to figure out 0:00:32.076 --> 0:00:36.916 how to implant electronic devices in people's brains, which is 0:00:36.956 --> 0:00:40.316 what he's doing, and in fact, as it happens, he's 0:00:40.316 --> 0:00:43.436 actually been preparing to do this kind of his whole life. 0:00:44.116 --> 0:00:45.916 You know, I sort of was born into the business. 0:00:46.476 --> 0:00:48.956 My dad is a neurologist who started on his career 0:00:48.996 --> 0:00:53.316 as an electrical engineer. You know, electrophysiology, clinical neuroscience and 0:00:54.036 --> 0:00:55.916 neurology and our surgery. I've been a part of my 0:00:56.236 --> 0:01:00.196 life forever as far as I can remember. And you know, 0:01:00.236 --> 0:01:02.956 brain computer interfaces the way we talk about them today 0:01:03.156 --> 0:01:07.956 didn't exist in the nineteen eighties, but the fundamentals were there, 0:01:07.996 --> 0:01:12.596 and so that's been percolating in some way forever. 0:01:18.956 --> 0:01:21.156 I'm Jacob Oldstein and this is What's Your Problem, the 0:01:21.156 --> 0:01:23.076 show where I talk to people who are trying to 0:01:23.156 --> 0:01:27.556 make technological progress. My guest today is Ben Rapaport. He's 0:01:27.636 --> 0:01:31.396 the co founder and chief science officer at Precision Neuroscience. 0:01:31.956 --> 0:01:34.916 Ben's problem is this, can you build a device that 0:01:34.996 --> 0:01:38.356 allows someone who is paralyzed to use a computer with 0:01:38.516 --> 0:01:42.116 only their thoughts and can you do it without sticking 0:01:42.156 --> 0:01:47.436 needles into their brain. Before he started Precision, Ben was 0:01:47.476 --> 0:01:50.796 a co founder of Neuralink. Neuralink is probably the best 0:01:50.876 --> 0:01:54.316 known brain computer interface company, and it was founded in 0:01:54.356 --> 0:01:58.156 twenty sixteen, right around the moment when modern AI was 0:01:58.276 --> 0:02:01.836 just emerging. And Ben told me the AI revolution was 0:02:01.876 --> 0:02:04.596 really what inspired the foundation of Neuralink. 0:02:05.396 --> 0:02:09.676 The kind of founding principles of neuralink were, you know, 0:02:09.956 --> 0:02:12.476 here's here's a point in time when we're thinking broadly 0:02:12.476 --> 0:02:15.516 about how the human brain is going to interact with 0:02:15.676 --> 0:02:20.156 artificial intelligence. And if breakthrough is an artificial intelligence are 0:02:20.196 --> 0:02:23.156 scaling at an exponential rate, you know, how's the human 0:02:23.156 --> 0:02:24.716 brain going to keep up with that? How are we 0:02:24.716 --> 0:02:28.276 going to keep communicating with artificial intelligence in a way 0:02:28.316 --> 0:02:30.196 that is feasible and productive. 0:02:30.516 --> 0:02:32.756 So that's a really different that's not how can we 0:02:32.916 --> 0:02:36.956 help people who are paralyzed. That's a much more sort 0:02:36.996 --> 0:02:39.796 of cognitive centric. It's about like the nature of human 0:02:39.876 --> 0:02:41.316 thought in the context of AI. 0:02:41.996 --> 0:02:45.156 So that was the that's kind of the raison deetra 0:02:45.236 --> 0:02:47.956 of neuralink, and it was a little different from a 0:02:48.516 --> 0:02:53.356 human focused medically oriented focus that Precision has taken. And 0:02:53.676 --> 0:02:57.236 these different focuses can you know, can and will coexist 0:02:57.276 --> 0:03:01.116 in an ecosystem in which multiple brain computer interface core 0:03:01.196 --> 0:03:04.556 technology has become widely available and are the standard to 0:03:04.556 --> 0:03:07.276 become the standard of care. But it became clear to 0:03:07.316 --> 0:03:10.276 me that that there was a need to also focus 0:03:10.876 --> 0:03:13.916 an effort within the community of bring computer interfaces on 0:03:14.436 --> 0:03:18.396 treating patients with untreatable diseases. That was the origin, you know, 0:03:18.516 --> 0:03:21.916 brain computer interfaces was really bringing this science and technology 0:03:21.996 --> 0:03:24.316 to a point where you know, people who today we 0:03:24.436 --> 0:03:27.756 think have has having really no treatment options, people with 0:03:28.156 --> 0:03:31.956 paralysis or an ability to speak, for example, from als, 0:03:32.436 --> 0:03:35.516 you know, really unlocking a world of possibilities for those people. 0:03:35.836 --> 0:03:39.876 But we really wanted to focus on those applications within 0:03:39.916 --> 0:03:43.916 bringing computer interfaces, and doing that, in my view, required 0:03:43.916 --> 0:03:47.516 making a few different design decisions than what we've made 0:03:47.716 --> 0:03:50.316 at Neuralink. You know, so those were the founding principles 0:03:50.356 --> 0:03:51.116 of Precision. 0:03:51.476 --> 0:03:54.796 You leave Neuralink to found Precision. Tell me tell me 0:03:54.836 --> 0:03:57.116 about what you you know, what you're setting out to 0:03:57.156 --> 0:03:59.516 create at Precision when you're launching the company, Like, what 0:03:59.636 --> 0:04:01.396 is it that you want to do and how is 0:04:01.436 --> 0:04:03.716 it different than what everybody else is doing. 0:04:03.756 --> 0:04:08.556 Yeah. The goal then and is today to build a safe, 0:04:09.156 --> 0:04:12.676 scalable brain computer interface that can become the standard of 0:04:12.716 --> 0:04:16.636 care in the treatment of patients people with a variety 0:04:16.636 --> 0:04:18.476 of diseases of the brain and nervous system that today 0:04:18.476 --> 0:04:22.156 are untreatable. That that includes various forms of paralysis and 0:04:22.396 --> 0:04:24.236 inability to communicate. 0:04:24.316 --> 0:04:26.836 And tell me about the tell me about the technology, like, 0:04:27.196 --> 0:04:29.196 tell me about the thing you're building and how it's 0:04:29.276 --> 0:04:31.436 different from what other people are building. 0:04:32.116 --> 0:04:34.716 Our philosophy has been that in order for a brain 0:04:34.796 --> 0:04:38.916 computer interface to really work in the real world and 0:04:38.996 --> 0:04:42.916 to unlock the potential of this technology for many millions 0:04:42.916 --> 0:04:46.236 of people. First, of course, it has to be incredibly safe. 0:04:46.436 --> 0:04:47.756 We see the use of the views of the term 0:04:47.756 --> 0:04:51.196 minimally invasive a lot, but really in my view, has 0:04:51.236 --> 0:04:52.196 to not damage the brain. 0:04:52.796 --> 0:04:54.756 So what does that mean in practice? 0:04:54.956 --> 0:04:58.156 Yeah, the tissue interface with the electrode involves kind of 0:04:58.196 --> 0:05:01.196 like little needles of the electrics are little needles and 0:05:01.236 --> 0:05:03.796 they penetrate into the brain. And there's been a lot 0:05:03.796 --> 0:05:06.316 of innovation in doing it, trying to do that very safely, 0:05:06.356 --> 0:05:09.916 But in my view, the most safe version of that 0:05:10.236 --> 0:05:12.716 is a version that just kind of caresses the brain 0:05:12.956 --> 0:05:16.076 but doesn't penetrate it. And it was at first thought, 0:05:16.236 --> 0:05:18.836 you know, certainly when we found a precision, many people 0:05:18.836 --> 0:05:21.356 thought that it was not possible to extract high quality 0:05:21.356 --> 0:05:24.516 signals from the brain without penetrating, and we and others 0:05:24.556 --> 0:05:26.836 have shown that, in fact, it's not only possible to 0:05:26.876 --> 0:05:30.076 do but has many advantages. So not that it's the 0:05:30.076 --> 0:05:32.996 only way or necessarily better or worse, but from the 0:05:33.036 --> 0:05:36.636 standpoint of people who have untreatable diseases and already have 0:05:36.716 --> 0:05:39.716 a very low threshold for damage to the brain, not 0:05:39.756 --> 0:05:42.636 doing any incremental damage to the brain for us is 0:05:42.756 --> 0:05:45.756 very very important. So that was sort of part one 0:05:46.476 --> 0:05:47.556 of precision. 0:05:47.276 --> 0:05:49.316 Is there before we get to part two? Is there 0:05:49.356 --> 0:05:52.076 a trade off? I mean, do you lose some amount 0:05:52.156 --> 0:05:55.676 of sensitivity or resolution? Is that the basic trade off? 0:05:55.836 --> 0:06:00.196 So we always get this question, you know, right, No, 0:06:00.276 --> 0:06:02.956 it's absolutely it's a good question, right, And so there's 0:06:03.076 --> 0:06:08.316 this false dichotomy. I think that more penetration into the 0:06:08.316 --> 0:06:12.236 brain equals higher quality signal, and if you don't do that, 0:06:12.356 --> 0:06:15.676 then you somehow sacrifice signal quality. But it's really not 0:06:15.836 --> 0:06:18.836 a one dimensional as one dimensional as that if you're 0:06:18.836 --> 0:06:22.036 a neuroscientist, then there's a trade off. If you care 0:06:22.076 --> 0:06:24.316 about recording from one neuron out of time, and you're 0:06:24.356 --> 0:06:27.196 studying the behavior of individual neurons, and you care about that, 0:06:27.556 --> 0:06:31.676 then you want what we called intracortical penetrating microelectrodes, the 0:06:31.676 --> 0:06:34.956 ones that can come up up close to an individual 0:06:34.996 --> 0:06:38.556 neuron and listen to those individual action potentials. And that's 0:06:38.596 --> 0:06:42.356 something that neuroscientists care about. So you don't want to 0:06:42.476 --> 0:06:46.956 use the same electrodes that we use for a precision 0:06:46.996 --> 0:06:50.476 But but if what you care about is is treating 0:06:50.516 --> 0:06:55.476 paralysis or sources of communication, what you care about is stable, 0:06:55.756 --> 0:06:59.316 high quality signals over a long period of time. And 0:06:59.356 --> 0:07:03.396 in that area, arguably, just based on the data, you know, 0:07:03.476 --> 0:07:06.356 the cortical surface electrodes that we use at precision are 0:07:06.476 --> 0:07:08.676 at least as good, if not better. And I think 0:07:08.716 --> 0:07:10.956 you know, I will tell because there's a few of 0:07:10.956 --> 0:07:12.836 these different systems that are now out there in the 0:07:12.836 --> 0:07:15.516 real world. What's really exciting is that this has come 0:07:15.516 --> 0:07:19.476 out of the laboratory, out of animal experiment territory, into 0:07:19.996 --> 0:07:23.996 human pilot clinical trials that we and neuralink and synchron 0:07:24.036 --> 0:07:27.556 and others are engaged and that's really where it's at. 0:07:27.756 --> 0:07:30.356 So tell me where you are now. I know you've 0:07:30.396 --> 0:07:34.156 done some amount of experimental work in people, right what 0:07:34.316 --> 0:07:35.716 is the frontier of your work right now? 0:07:35.996 --> 0:07:40.036 Yeah, we've now implanted our electrode raise in almost thirty 0:07:40.316 --> 0:07:44.476 patients over the last two years. These are pilot studies 0:07:44.836 --> 0:07:47.396 across four major medical centers and the US that are 0:07:47.436 --> 0:07:50.556 partnering with US, and all of those studies are really 0:07:51.036 --> 0:07:54.036 they're temporary placements of the electrodes. So there are studies 0:07:54.076 --> 0:07:57.636 that are run in nations who have volunteered to have 0:07:57.716 --> 0:08:01.756 the electrodes placed alongside clinical standard electrodes as part of 0:08:01.796 --> 0:08:05.516 a under resurgical procedure that they're already undergoing. And we've 0:08:05.556 --> 0:08:09.476 been using those opportunities to basically validate the quality of 0:08:09.476 --> 0:08:12.956 electroctavity that we can record on those electrodes and to 0:08:12.996 --> 0:08:18.636 demonstrate that our algorithms can in fact basically decode intention 0:08:18.956 --> 0:08:22.676 and thought as intended by health essentially healthy volunteers. 0:08:23.116 --> 0:08:28.036 So the array itself, like, what does it look like? 0:08:29.716 --> 0:08:31.916 So the brain lives in the skull, so it's a 0:08:32.196 --> 0:08:37.596 it is a soft tissue that's kind of jelly like inconsistency, 0:08:38.156 --> 0:08:41.716 and so the best way to generally interface with it 0:08:41.756 --> 0:08:45.356 is with something also that is soft and flexible. And 0:08:45.396 --> 0:08:47.556 the surface of the brain is many of us have 0:08:47.596 --> 0:08:51.636 seen in pictures, is curved or undulating, and so our 0:08:51.676 --> 0:08:56.676 electrode array is a thin polymer that's many times thinner 0:08:56.676 --> 0:08:59.476 even than a human hair. So it's a film, and 0:08:59.516 --> 0:09:04.396 embedded in that film are tiny little dots of platinum, 0:09:04.916 --> 0:09:08.956 each one connected to it very very very thin platinum wire. 0:09:09.556 --> 0:09:12.916 And so that film with the tiny little dots of 0:09:12.956 --> 0:09:16.956 platinum inside, can be placed over the brain surface and 0:09:16.996 --> 0:09:21.956 it conforms to that curved surface. So each of those 0:09:21.996 --> 0:09:25.516 little platinum electrodes touches the surface of the brain at 0:09:25.516 --> 0:09:28.116 a very discreete point, and so it can record the 0:09:28.156 --> 0:09:31.676 electrical activity from the area of the brain just under 0:09:31.796 --> 0:09:32.956 that its touching basically. 0:09:33.876 --> 0:09:38.316 Okay, So in these trials, you put this implant on 0:09:38.356 --> 0:09:41.116 a patient's brain and then what. 0:09:42.156 --> 0:09:45.116 So let me describe maybe one of the paradigms that 0:09:45.116 --> 0:09:48.236 we use at one of our partner sites. So ian 0:09:48.316 --> 0:09:52.036 Cahegus is the neurosurgeon at Penn who's our partner, and 0:09:52.196 --> 0:09:54.516 he is a surgeon who specializes in the treatment of 0:09:54.556 --> 0:09:58.076 Parkinson's disease. One of the ways of treating Parkinson's disease 0:09:58.356 --> 0:10:01.796 is a procedure called deep brain stimulation, in which electrodes 0:10:01.836 --> 0:10:05.476 are placed deep within the brain to stimulate those areas 0:10:05.596 --> 0:10:09.516 that are responsible for modulating the tremor. Doctor ca Vegas, 0:10:09.556 --> 0:10:13.076 among many others, performs these procedures at least a part 0:10:13.076 --> 0:10:16.036 of them awake in order to make sure effectively that 0:10:16.036 --> 0:10:18.916 the exact right place is being targeted and the brain 0:10:18.956 --> 0:10:22.476 doesn't feel pain. And so it's not only possible but 0:10:22.516 --> 0:10:25.356 beneficial to do these procedures at least partially awake. So 0:10:25.636 --> 0:10:28.996 in those procedures, we take a basically a fifteen minute 0:10:29.036 --> 0:10:34.316 window and doctor Ahigis places the precisional electrode directly over 0:10:34.396 --> 0:10:37.276 the motor cortex, a portion of the motor cortex that 0:10:37.276 --> 0:10:41.556 controls hand movement. And this has provided for us and 0:10:41.596 --> 0:10:46.396 for the community, the highest resolution picture of the human 0:10:46.436 --> 0:10:50.156 motor cortex and the awake human ever in the history 0:10:50.196 --> 0:10:53.076 of the world. So you know, the area of the 0:10:53.076 --> 0:10:55.916 brain of the motor cortex that controls hand movement is 0:10:55.956 --> 0:11:00.956 about the size of a postage stamp. And critically to 0:11:01.036 --> 0:11:06.116 understand is the neurons that are responsible for coordinating those movements. 0:11:06.796 --> 0:11:10.116 They all live within a two milimeters your layer of 0:11:10.716 --> 0:11:14.476 tissue that's just at the surface of the brain. So 0:11:14.596 --> 0:11:17.796 all that critical computation and activity is happening very very 0:11:17.796 --> 0:11:20.636 close to the surface. And so it's good. 0:11:20.396 --> 0:11:24.076 For you, good for us method. So so what actually happens. 0:11:24.076 --> 0:11:26.916 So you have a patient who's there, you put your 0:11:27.236 --> 0:11:30.636 array on the on the portion of their motor cortex 0:11:30.676 --> 0:11:34.356 that controls hand movements, and then you say, wiggle your finger. 0:11:34.276 --> 0:11:37.836 Exactly, so we say, we say, we say, basically, you know, 0:11:37.916 --> 0:11:40.196 we asked. We walk the patient through making a certain 0:11:40.236 --> 0:11:43.356 number of gestures, you know, open hand, close hand, make 0:11:43.396 --> 0:11:47.276 a peace sign, and we watch, and I say metaphorically watch, 0:11:47.316 --> 0:11:49.676 We watch what the patient is doing, and we watch 0:11:49.716 --> 0:11:52.236 what is happening on the surface of the brain. And 0:11:52.236 --> 0:11:55.316 and here is you know, where modern machine learning plays 0:11:55.316 --> 0:11:59.396 a tremendous role, because this exactly, you know, this is 0:11:59.436 --> 0:12:01.716 the AI portion of it, because this is this is 0:12:02.236 --> 0:12:04.916 the so called training data. So this this is a 0:12:04.956 --> 0:12:09.516 calibration phase in which our algorithms learn what the brain's 0:12:09.636 --> 0:12:12.116 signals to the hand look like in a given patient. 0:12:12.796 --> 0:12:16.876 So there's a characteristic signature, electrical signature that happens in 0:12:16.916 --> 0:12:19.876 the moments before an action is done, and it's a 0:12:19.916 --> 0:12:23.676 little bit different in each person, and learning that signature 0:12:23.716 --> 0:12:27.396 for that person allows us to recognize when the brain 0:12:27.476 --> 0:12:29.556 is telling the hand to make a particular gesture, when 0:12:29.556 --> 0:12:31.476 the fingers are supposed to move in a particular way, 0:12:31.476 --> 0:12:34.476 when the hand opens and closes, and after about three 0:12:34.556 --> 0:12:37.916 to five minutes of training, we then have a trained 0:12:37.956 --> 0:12:42.076 algorithm that can recognize not just movement, but the intention 0:12:42.156 --> 0:12:44.876 to move. And so we then use the balance of 0:12:44.956 --> 0:12:47.236 the time that we have with those patients to ask 0:12:47.556 --> 0:12:50.596 the patient to move and validate that we're predicting the 0:12:50.636 --> 0:12:55.396 correct movement, and then to imagine movement without moving, and 0:12:55.436 --> 0:12:59.436 that too we can accurately predict. And so these procedures 0:12:59.716 --> 0:13:05.196 become the basically healthy volunteer test bed for patients who 0:13:05.556 --> 0:13:08.796 can't actually move, the paralyzed patients that we'll be treating 0:13:09.156 --> 0:13:11.316 within the next couple of years. So that's that's the 0:13:11.436 --> 0:13:14.556 nature of this first phase of pilot trials. 0:13:14.876 --> 0:13:19.036 You mentioned that each person is different in terms of 0:13:19.076 --> 0:13:25.196 the patterns of neuron activity for each hand motion. In 0:13:25.236 --> 0:13:28.396 this context, how different is it like kind of like 0:13:28.436 --> 0:13:30.916 a Southern accent versus a New York accent. Is it 0:13:30.996 --> 0:13:33.756 like an entirely different language if that kind of metaphor work. 0:13:33.836 --> 0:13:36.796 Yeah, No, that's a perfect metaphor. And it's it's kind 0:13:36.796 --> 0:13:39.516 of like that. So you know, you if you're trying 0:13:39.556 --> 0:13:41.836 to learn a new language or a dialect, you know 0:13:42.036 --> 0:13:45.396 that there are words, and you know that they're spoken 0:13:45.396 --> 0:13:47.596 in a particular frequency range, so you kind of know 0:13:47.596 --> 0:13:49.916 what to listen for, and you kind of know the cadence. 0:13:49.996 --> 0:13:51.716 So when there's a word, you know that's a word, 0:13:52.116 --> 0:13:54.716 but you might not know what it means until you 0:13:54.916 --> 0:13:57.796 listen in to conversation and you've seen the context. 0:13:58.036 --> 0:14:01.396 So so like pretty different. Like it wouldn't work to 0:14:01.516 --> 0:14:05.916 just make a generic algorithm and put her on my brain. 0:14:05.836 --> 0:14:08.836 Because it doesn't work to make a generic algorithm. But 0:14:09.396 --> 0:14:11.716 that's an area where there's been a lot of just 0:14:11.756 --> 0:14:15.676 fascinating development. And so a good example of this is, 0:14:15.876 --> 0:14:18.076 you know, Siri works out of the box for most 0:14:18.116 --> 0:14:20.876 people pretty well, right. 0:14:21.076 --> 0:14:23.516 Talk into your eye, right, it works. 0:14:23.316 --> 0:14:25.956 Right, It works pretty well, and then you need to 0:14:25.996 --> 0:14:28.236 train it to make it better, and then it listens 0:14:28.236 --> 0:14:31.116 to you in the background and gets even better. And 0:14:31.156 --> 0:14:32.996 so that that's a good that's a good analogy. So 0:14:33.316 --> 0:14:36.196 it is possible for us to build, you know, a 0:14:36.196 --> 0:14:40.876 translation algorithm that works somewhat out of the box, but 0:14:40.996 --> 0:14:45.036 we build into it a calibration phase that knows something 0:14:45.076 --> 0:14:48.996 about the structure of brain signals and how they interact 0:14:48.996 --> 0:14:53.676 with and relate to movement or speech. And that's what 0:14:53.796 --> 0:14:57.036 basically allows us to use only relatively small amounts of 0:14:57.076 --> 0:14:59.116 calibration data. I mean, you know, we can do a 0:14:59.156 --> 0:15:01.116 lot with a small amount of calibration data. 0:15:01.196 --> 0:15:05.836 So you're doing a sort of pilot study. Now, when 0:15:06.516 --> 0:15:07.836 what's the next big step? 0:15:08.756 --> 0:15:11.676 So I want to be careful about what I say 0:15:11.996 --> 0:15:15.756 before it happens. But we do anticipate being able to 0:15:16.156 --> 0:15:19.476 in the very near future extend what are now short 0:15:19.556 --> 0:15:24.156 duration file with studies that last only the span of 0:15:24.196 --> 0:15:25.916 time that we have access to the brain within a 0:15:26.236 --> 0:15:30.636 standard androsurgical procedure, which is relatively short. We anticipate having 0:15:30.676 --> 0:15:34.916 ways of extending that with regulatory approval, to hopefully many 0:15:34.956 --> 0:15:37.956 days and weeks within the calendar year. And then, of course, 0:15:37.996 --> 0:15:41.316 this is all in the service of permanent implants that 0:15:41.636 --> 0:15:45.476 wirelessly communicate with the outside world, and that will be 0:15:45.636 --> 0:15:48.116 the basis of our pivotal clinical trial a couple of 0:15:48.156 --> 0:15:48.876 years hence. 0:15:51.316 --> 0:15:53.796 Still to come. On the show, Ben and I discussed 0:15:53.836 --> 0:15:57.916 the possibility of using brain computer interfaces in healthy people, 0:15:58.716 --> 0:16:09.756 also the meaning of consciousness. Just before the break, Ben 0:16:09.796 --> 0:16:12.996 mentioned that pivotal clinical trial that they're building up to, 0:16:13.156 --> 0:16:15.236 and so I asked him what exactly they're going to 0:16:15.276 --> 0:16:16.716 be doing in that trial. 0:16:17.116 --> 0:16:19.556 So the first clinical application is going to be for 0:16:19.556 --> 0:16:23.476 the treatment of severe paralysis, okay, And the device will 0:16:23.516 --> 0:16:26.516 be an implant that has the electrodes on the brain 0:16:26.716 --> 0:16:29.756 and an implant within the chestwell that provides power and 0:16:29.836 --> 0:16:33.156 data transfer to the outside world to communicate with the 0:16:33.156 --> 0:16:36.396 external devices like a computer. And that system will allow, 0:16:36.476 --> 0:16:39.156 for example, a person with a spinal cord injury really 0:16:39.156 --> 0:16:41.756 to hold the desk job that will allow them to 0:16:41.916 --> 0:16:46.556 operate effectively a word processing program, email, serve the internet, 0:16:46.876 --> 0:16:50.036 have a zoom conversation, operate and expel a sales spreadsheet, 0:16:50.156 --> 0:16:53.916 use PowerPoint, have the ability to re enter the workforce 0:16:54.276 --> 0:16:57.796 with a level of personal and economic self sufficiency that 0:16:58.076 --> 0:17:01.196 allows them to, you know, certain freedoms that they don't have, 0:17:01.356 --> 0:17:04.556 and that our core to being a part of modern society. 0:17:04.876 --> 0:17:07.796 That is, for us a major goal. Number one, I'm 0:17:07.836 --> 0:17:12.916 quite sure that past the technolog becomes provenly safe and effective, 0:17:13.396 --> 0:17:17.076 that other disorders and conditions that are perhaps less dramatic, 0:17:17.476 --> 0:17:20.236 you know, will benefit from this and in other forms 0:17:20.276 --> 0:17:23.916 of technology. And part three is there's a lot that 0:17:23.956 --> 0:17:26.116 I'm sure that we're not even imagining right now. You know, 0:17:26.356 --> 0:17:30.076 the brain computer interface, at least the precision system is 0:17:30.116 --> 0:17:33.476 really in some ways a platform technology because it's it 0:17:33.636 --> 0:17:37.836 translates the wet and difficult to access, delicate, you know, 0:17:37.876 --> 0:17:43.676 biological signals of the brain into robust digital bitstreams and 0:17:43.756 --> 0:17:46.356 allows us to compute on them in a scalable way. 0:17:46.676 --> 0:17:48.796 The brain computer interface is not a substitute for a 0:17:48.836 --> 0:17:51.436 keyboard in a mouse. It's not a substitute for a 0:17:51.556 --> 0:17:56.156 gestural interface or a voice interface. It's another kind of 0:17:56.196 --> 0:17:59.436 interface with the brain, just like it was would have 0:17:59.476 --> 0:18:01.996 been impossible to predict based on the keyboard, a loan, 0:18:02.156 --> 0:18:05.476 or the graphical user interface alone, all of the different 0:18:05.516 --> 0:18:08.516 applications that have emerged. I think, as long as we 0:18:08.596 --> 0:18:13.356 build a safe, reliable interface and make that responsibly available 0:18:13.756 --> 0:18:16.836 kind of skuy's the limit. And I can't even hazard 0:18:16.876 --> 0:18:18.716 to guess at some of the things that will come next. 0:18:18.756 --> 0:18:20.956 So I think there's a there's a whole generation of 0:18:21.116 --> 0:18:23.996 discovery and innovation waiting to happen after we get this 0:18:24.196 --> 0:18:27.276 across the line into patients to become standard of care. 0:18:27.956 --> 0:18:30.756 Could you imagine it being used in healthy people for 0:18:31.716 --> 0:18:35.476 you know, the computer and the brain application. 0:18:35.836 --> 0:18:37.796 Yeah, I could eventually, in a sense, I would love 0:18:37.796 --> 0:18:38.516 that to be the case. 0:18:38.876 --> 0:18:42.476 I think I'm ambivalent about that one. Tell me why 0:18:42.476 --> 0:18:43.196 you'd love that to be. 0:18:43.156 --> 0:18:45.636 The case, Well, because it will have meant that we've. 0:18:46.436 --> 0:18:48.876 Well, yes, it'll mean your thing works really well, it 0:18:48.956 --> 0:18:49.956 is wildly safe. 0:18:50.036 --> 0:18:52.556 Yes, that's true, right, Yeah, So we build with that 0:18:52.636 --> 0:18:55.156 goal in mind in a way, right, just because in 0:18:55.276 --> 0:18:58.436 order for something to be accepted by an eble body 0:18:58.596 --> 0:19:03.316 person who has zero risk tolerance, right and basically only downside, 0:19:03.436 --> 0:19:06.596 if something goes wrong or doesn't work properly, you need 0:19:06.596 --> 0:19:09.996 it to work just in a bulletproof way. That's that's 0:19:09.996 --> 0:19:12.676 the kind of system that we're trying to engineer. 0:19:12.796 --> 0:19:15.836 Yes, from that point of view, it makes perfect sense. 0:19:15.876 --> 0:19:18.556 Then if that is true, then then you have built 0:19:18.556 --> 0:19:22.196 a wildly safe and effective device exactly. 0:19:22.236 --> 0:19:23.916 So, if you and I were having this conversation and 0:19:23.916 --> 0:19:25.676 you said to me, gosh, I would love to write. 0:19:25.716 --> 0:19:27.876 I mean that would mean that all those doubts had 0:19:27.916 --> 0:19:31.276 been erased. And in order to erase those doubts, we 0:19:31.396 --> 0:19:33.476 have to prove certain things to the world, and that's 0:19:33.516 --> 0:19:34.636 that's really our job. 0:19:35.756 --> 0:19:39.276 Would you would you want if you were healthy? Would 0:19:39.316 --> 0:19:41.596 you want to have your device in your brain? If 0:19:41.636 --> 0:19:43.796 it were safe and effective? 0:19:43.956 --> 0:19:46.436 I would have to do certain things that that the 0:19:46.476 --> 0:19:49.196 device can't do yet, yeah, rush, but I wouldn't definitely 0:19:49.196 --> 0:19:50.596 wouldn't rerule it out when we get there. And I 0:19:50.596 --> 0:19:53.756 mean it's like sometimes with technology, it's it's hard to 0:19:53.796 --> 0:19:55.916 wrap your mind around what's going to happen in a generation, 0:19:56.156 --> 0:19:59.476 right too little kids and we're always talking about like 0:19:59.676 --> 0:20:01.916 should the kids actually get to use an iPhone? 0:20:02.356 --> 0:20:05.156 Hold out for as long as you can so, right, 0:20:05.196 --> 0:20:07.876 because so it's not exactly a choice, right, that's the 0:20:07.996 --> 0:20:10.116 that's the thing you think, like, oh an iPhone? 0:20:10.156 --> 0:20:12.876 That's my point. By the way, I'm very very permissive 0:20:13.556 --> 0:20:14.196 and me too. 0:20:15.516 --> 0:20:17.916 You know what finished me was COVID, Like we held 0:20:17.916 --> 0:20:21.396 out really strong and then COVID hit did ason. 0:20:21.636 --> 0:20:23.476 So but the reason I bring that up is that, 0:20:23.516 --> 0:20:26.676 you know, like our parents could not even have conceived 0:20:26.836 --> 0:20:28.796 of even that question, right. 0:20:28.756 --> 0:20:30.956 Yes, But I mean the other way to think about 0:20:30.996 --> 0:20:34.236 that is, like, you know, I'm pro progress and pro technology, 0:20:34.396 --> 0:20:39.316 but like having kids makes me wish iPhones didn't exist, right, 0:20:39.436 --> 0:20:41.676 makes me wish like, sure, give them a flip phone 0:20:41.716 --> 0:20:43.556 so they can text their friends that call me if 0:20:43.596 --> 0:20:46.156 something goes wrong. But well, you know, I don't know. 0:20:46.356 --> 0:20:48.556 But on the other hand, I make podcasts for a little. 0:20:49.236 --> 0:20:52.396 It's an interesting discussion, right, and you know, we sometimes joke, 0:20:52.476 --> 0:20:54.916 but somehow kids are born now knowing how to swipe 0:20:54.956 --> 0:20:58.796 and navigate the phone interface. Right. So my point is 0:20:58.836 --> 0:21:01.836 that in twenty years it's going to be a different conversation. 0:21:02.196 --> 0:21:04.676 There's a lot of kids of people in the company, 0:21:04.756 --> 0:21:07.356 and they know what we're doing, you know, like girls 0:21:07.396 --> 0:21:09.796 know what we're doing, and their view and the technology 0:21:09.836 --> 0:21:13.596 is different. They see it as something that exists, and 0:21:13.636 --> 0:21:15.396 when you're bored into it, you have kind of a 0:21:15.396 --> 0:21:18.236 different sense of what's okay and what's normal. And that's 0:21:18.276 --> 0:21:20.996 the generation that's growing up today is going to grow 0:21:21.076 --> 0:21:24.156 up with bring computer interfaces just being a normal thing. 0:21:24.916 --> 0:21:28.436 Yeah, maybe your grandkids will feel about bring computer interfaces 0:21:28.476 --> 0:21:30.076 the way your kids feel about iPhones. 0:21:30.236 --> 0:21:31.516 It's going to happen faster than that. 0:21:36.356 --> 0:21:38.436 We'll be back in a minute with the lightning rim 0:21:48.876 --> 0:21:53.956 tell me about the metabolic factors limiting performance in marathon runners. 0:21:54.556 --> 0:21:58.756 Okay, right, so that was a paper that I wrote 0:21:58.916 --> 0:22:01.356 now more than a decade ago. So I'm a dedicated 0:22:01.396 --> 0:22:05.156 marathon runner and I run forty something marathons over twenty 0:22:05.396 --> 0:22:07.916 plus years. There's a longer story, which we don't have 0:22:07.956 --> 0:22:10.196 time for now, is to I wrote that paper. 0:22:10.236 --> 0:22:12.716 Give me what's the short version of that story of 0:22:12.716 --> 0:22:13.596 why you wrote the paper. 0:22:13.636 --> 0:22:16.396 The short version is it shouldn't be metabolically possible to 0:22:16.476 --> 0:22:20.236 run a marathon because interest everybody. Everybody thinks the paradox 0:22:20.316 --> 0:22:22.596 is that you know, you can't eat enough pasta to 0:22:22.636 --> 0:22:23.956 get through twenty six miles. 0:22:24.596 --> 0:22:26.996 Uh, just like if you do the math, there's not 0:22:27.196 --> 0:22:29.236 enough energy stored in the body. 0:22:28.956 --> 0:22:30.636 If you do the simple math. There seems to be 0:22:30.636 --> 0:22:32.996 a paradox that you can't you can't eat enough pasta 0:22:33.036 --> 0:22:34.676