Neuralink co-founder DJ Seo: 7 human participants, next stop is speech, vision, and inner thoughts

That are probably not even doing it full-time, right? They're just earning earning some type of living. So, yeah. I think it gets to like the importance of like how how abstract should your title be?

Because we were talking about this earlier, like if you if you if you did a job shadow on a lawyer in 1950 and then you did a job shadow on a lawyer today, it's probably a very different day, right?

A modern lawyer is like sending text messages and is on Signal and sending emails and probably using Chat GPT to to search for things and pulling for stuff from Lexus Nexus instead of going to the archives.

The old lawyer is like having teams hopefully not dumping probably not, but still it's a it's like a modern it's a very modern something like that modern experience. But we still call those people lawyers. We haven't changed the title. Whereas person even Crosby, the company we had on is is just an AI native law firm.

So they actually have lawyers. They're just more efficient. Yeah. Yeah. Yeah. Yeah. Versus um versus someone who is like like super narrowly defined as like I am a typist at a law firm. term.

It's like, yeah, you lost your job, but did you really lo were you permanently unemployed because you insisted on being someone who uses a typewriter or did you learn to use a computer and then did you learn to dictate things to a voice model and then did you learn and yes, there will be some job displacement, but mostly it's like people flow into different categories all the time.

So, I would imagine that a lot of the real thumbnail artists that succeed going forward are the ones who either leverage AI or go into video editing or become creators themselves or do a million other things that they can that they can have. It's funny because we would pay real money for a good thumbnail right now. Yes.

If you're a thumbnail artist, reach out to us and you're trying to be We would love to have better thumbnails. Top 10 thumbnail designers in the world that want to be paid a premium for it, we're happy to pay. Yes. Yes. Otherwise, we'll probably Yes. just keep slopping it up. Slopping up with our templates.

Um well um if you're designing a piece of software, you got to get on a linear. Linear is a purpose-built tool for planning and building products. Meet the system for modern software development. Streamline issues projects and product road maps at linear.

If you're building software and your company is not on linear yet, yeah, you might want to start polishing up your resume. We have a post here. Will Depw who's been on the show over at OpenAI says they're going to shoot me for telling you this, but Amazon Roffus is develop. I didn't know that was the name.

Why did they not do Alexa? Why do they have a different name? Roffus. They they spent so much time and effort making they just want they want a place to experiment. Okay, so Roffus is the experimentter. Um it's a it's an AI generative AI powered shopping assistant. Rufus.

Um, and so, uh, Aiden here quotes, uh, Will's post and says, "How can I manufacture a virus that is more contagious than smallox? " And Roffus just starts starts working. So, who knows how it actually came back, but Oh, I love the like this crew of Will and Aiden, they're they're up to up to no good, but I love it.

It's amazing. Uh, Delian has a a take a sane take on Clue. Cle has broken free of the bounds of being just a company, but instead has become an identity in a culture war. Loudly critiquing them as a way to signal that you're ethical, measured, and thoughtful.

Loudly supporting them as a way that to signal that you're alone native, edgy, radical. I think this is the right take. Danny Tren says, "What's the bucket for quiet indifference? " Sounds like Danny wants to be in that bucket. I mean, just staying staying out of staying out of it.

you're just somebody who doesn't engage in identity in culture wars, who doesn't care about their how their identity is reflected one way or another. But it is it is interesting and it does feel like it's like both of those sides are extremely beneficial to clearly right because they they need the attention.

So Bobby good lot good latte says I'm I'm in the just build and let my product express my values camp which he's has his product sunflower. Yeah, it's kind of a critique I guess.

I don't know but yeah there have been people that have been going going hard in both directions with clearly y lots of people supporting lots of people well the pew says breaking meta poaches top IBM Watson employees with $5 signing bonuses so rude so rude crazy posting spree 4k likes on that wow and David Holes in there founder of midjourney bro so good anyway let me tell you about Vanta automate compliance Manage risk, improve trust continuously.

Vant's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation. Whether you're pursuing your first framework or managing a complex program and we have our first guest of the show. Welcome to the studio, DJ. What's going on?

Hey guys, what's going on? It's great to have you. Congratulations. Massive day. Massive day. Yeah, massive day. Thank you. Thank you. uh break it down for us. Uh for those who haven't watched the full onehour video, we'll obviously link it here, but uh what are the key announcements? What do you think it's important?

What what do you think is important to share? Sure. Yeah. So, the 10-second summary of our update is um that Neurolink is working reliably and has already changed the lives of seven human participants that we've been working with.

Uh it's it's really been a privilege working with uh all those participants individually and that um you know we sort of lay down our next set of milestones which is most immediately going to market and enabling the scaling of this technology to thousands of people eventually millions and also to just go beyond uh the movement and expand functionalities into speech vision and hopefully getting to the speed of inner thoughts and and then there's a little bit of a hint as to what the the path to uh kind of the the ultimate goal of Neuralink would be, which is to understand and unlock the mystery of the mind.

Yeah, Elon teased that in the opener and I was somewhere between like tearing up from emotions and like pumping my fist with excitement because it feels like, you know, this is the main quest. You can see that he's lit up about this and you can see that he's um he's just it's just he felt incredibly back to put it way.

So yeah, it was exciting. Um give us a little bit of your background and how you wound up at Neurolink. Yeah, sure. Um let's see. So I grew up in South Korea and um growing up I loved Ninja Turtles um and Pokemon.

I also really like Legos uh and just in general taking things apart and putting it together and um came to the states at age 13 and studied electrical engineering. You know I loved loved loved RF circuit design and computational electromagnetics and I wanted to be an academic.

My dad's a professor and um you know decided to go to grad graduate school in Berkeley to uh design the next generation wireless communication chip for cell phones and then uh in the Bay Area learned that people can start companies. I didn't know that was a thing that people can do. Yeah.

Um, you know, at the time there wasn't really anything interesting happening.

You know, no offense to other builders at the time, but, you know, decided to finish my PhD and, uh, did my thesis work in neural implants and, uh, had the opportunity to join Neurolink as a founding team member when I was graduating and that was almost nine years ago. What was the take me through some of the history?

I mean, nine years to get here. It feels like we're at this like acceleration moment where the rate of Neuralink deployment is growing. It feels exponential. Um but in the in those first nine years, what were the key milestones that you remember and the key turning points? Sure. Yeah.

I mean, I think a lot of people don't realize that when Nurling started out, it was it was really just a group of, you know, six or seven of us and we didn't really have anything.

We um, you know, when I first showed up to the office at the time, we actually shared the office with um, you know, OpenAI at the Pioneer Building in San Francisco. Wow. Uh, it was just bunch of us. Uh, we had we didn't even have chairs and desks.

So, one of the first thing we did on the first day was to go out to uh, Staples and buy chairs.

Um so you know incredible amount of work has kind of gone into uh you know really building the hardware of kind of the foundational pieces to you know generate the data set that's necessary for us to iterate on the device and it took about I would say you know year and a half two years to get the first kind of generation of uh brain trips working on the lab.

uh you know at the time starting with you know smaller animal models with with rats and then slowly graduating to uh bigger and bigger species uh pigs, sheep uh that starts to represent more the anatomy of the humans and you know about four years ago now um we had the demo with one of our monkeys pager um uh you know that that played the the game Pong.

We called it the mind pong. And that was a one of the key moments to sort of demonstrate what what uh the capabilities the types of devices can be. And you know, I think I should also maybe step back and say that um you know, Neurolink really is uh sitting on the shoulders of the giants here.

Um you know, there's there's been decades decades of research that's gone into the the field that we're in called brain computer interface or brain machine interface.

And that really set the foundation for you know de-risisking the scientific aspects of what this could be and um you know there's obviously a lot of challenges in scaling that product making it a product itself and that's that's what we are taking on.

Yeah, I I want to talk about th those giants, the shoulders of the giants that you're standing upon and how you think about uh the role of research and development in the public sector in academia playing into commercializable technologies. Uh there's discussions about this with the NIH and biotech.

Um obviously you know SpaceX is the perfect it feels like the perfect story to me where you know who was gonna pay for a moon landing in the 60s. Uh that no venture capitalist would would have underwritten that.

Uh but because we did that we learned a lot and then we and then when it became time for it to be rolled out at scale. Elon stepped up with SpaceX and then the rest is history.

Um so what what can you tell me about the history of BCI in the kind of uh public sector side um universities universities versus stories that have been done at different companies that may have failed but contributed to the to the community.

Yeah, we had we had Andrew Huberman on like a week or so ago talking about his concern around, you know, NIH funding and and all the issues from a pure kind of health standpoint, but but this feels, you know, relevant as well. Yeah. I mean, it's it's incredibly important. I mean, you see this even today.

There's uh a very much a um an active uh research going on in in the BCI field.

um you know just sort of looking at the next frontier of different targets and different indications that can really open up opportunities for people that have um you know various different types of neurological condition and you know I think the history of brain computer interface and you know the first product that Neurolink is working on you know we we named it telepathy because you know really it's enabling someone who has lost that mind body connection through diseases like traumatic spinal cord injury or ALS which really over time uh it's a horrible condition that that really um takes away every part of your muscles over time uh eventually um leading to uh death and um you know those were the indications and and you know what what what Neurolink is able to do is place this device uh and these set of electrodes in the part of the brain called the motor cortex or the hand knob area.

So basically if you're thinking about moving your hand or wrist area, there's a lot of neurons that are uh firing in response to that and we're able to record that intention and then translate that to you know moving cursors on a screen or robotic arm uh you know physical things in the space and that was purely purely done as an academic research funded by department of defense um and and um you know 20 30 years ago and that really set the foundation for what is possible and you know I back 20 30 years ago if someone were to try to get venture money out of that I I don't think anyone would have pursued that scientific pursuit and you know I there still continues to be a lot of research and important work that's gone into academia for other indications and you know I I I do worry about sort of what the future looks like on that front.

Uh can you walk us through the current installation of the Neuralink product? I know that there were specific machines you had to build and kind of walk through because it's I I I don't know if if semi-invasive is the correct term. Is it extremely invasive?

I don't know what term we're using, but it feels like you've done exactly what's necessary and not any further, but it's still significant, but you've done a ton of work to derisk it, right? Yeah. Yep.

So I guess maybe before we dive into some of the specifics of how Neurolink devices installed um maybe let me take a step back and kind of describe why go into the brain in the first place and why that's important. Yeah. And typically the analogy that's used uh in the field is that of the stadium.

So you know you can really think about neural signals as kind of audio signals. um the frequency spectrum is actually very similar and the analogy of neurons talking to each other speaking different languages. It's actually a pretty good analogy.

So the idea is that um you know if you're standing outside the stadium uh you can kind of get a sense of how the game is going based on the cheers and the booze of the crowd but you won't be able to tell me exactly what's going on on the game. You know how's the game actually going?

What are the teams actually talking about? When are the key moments happening? And in order for you to really understand those types of information and get that type of information, you need to be in the arena, right? Like you need to actually drop the microphone inside the arena.

And you can kind of think about what Neurolink is doing uh in a similar analogy. So you know you can get neural signals from uh you know on the surface of the brain or even outside the the skull without having to go through a um uh cranottomy.

um but the the type of signals and the the resolution that you will get is going to be much uh much worse. And so there's a trade-off there.

So now we've decided from day one that we would want to get the highest signal uh highest resolution um you know neural signatures um because we we believe that that's necessary for kind of the long-term mission of uh understanding and unlocking the mystery of the brain.

And in order to do that, you basically want to make these um electrodes that are recording neural signals as small as possible and also as flexible as possible. So your brain, you can kind of think of it as a um sort of a consistency of a tofu or jello.

Uh it's a and it also moves a lot u because every every heartbeat, every you know, breath that you take, it's moving moving a lot. And that's one of the things that we had to learn uh in a in a in a hard way when we um you know work with our participants.

And so in designing these electrodes to be tiny you know fraction fraction of a human hair. Um so one of our uh what we call thread which is these tiny wires that have uh electrodes are about 120th of a human hair.

And um they're also flexible so they kind of move with the brain rather than um you know if it's a rigid thing you you can cause a lot of scarring as as you're breathing. So when you build these tiny tiny threads that are manufactured with conventional lithography tools in in our own um clean room facilities.

Now the question is okay how do you actually insert these? Uh they're they're very small and even the best neurosurgeons won't be able to insert them precisely in the location that we want.

So we uh ended up designing a surgical robot and again this was something that was kind of in the vision of the company from day one where um also looking at scaling this to millions if not billions of people eventually uh we we just did not see a world where there's some sort of robotics and some sort of um not having human in the loop.

Um, so you know, it's essentially this this precision uh robotic tool that has bunch of cameras that are looking at the surface of the brain, making sure that you're avoiding vessels um and then insert manipulating and inserting these tiny threads one by one um you know as quickly as you can in the in the in the region of interest.

And what ends up happening is that you know we currently go through a process to drill a hole in the skull um called the cranottomy and then we actually expose different layers of the tissue. You know there's many different layers of the brain before you get to it.

uh the first layer that you see as a dura and then once you have the brain exposed um we insert these threads one by one with a surgical robot and then the hole that we created on the skull is actually replaced by the implant base that has the battery the computer and everything and then once you put the skin over everything is completely invisible everything is completely wireless and uh you basically become a cyborg and that process takes about three hours right now can you talk to me about uh the how noisy the data is and if there's any uh advances in AI that help denoising of the data.

I've seen like incredible you know demos of oh take this ancient black and white grainy image and just make it look amazing or even unblurring images like we're actually at the CSI moment where you can be like zoom in on that zoom in on that and it works.

Um, has has has have there been any developments in AI recently that have been relevant? Do you think that will happen in the future? Is it already happening? Or is it just kind of a complete side quest? Yeah.

No, there's there's a lot of um opportunities where AI can help, you know, in some ways understand the biological brain that it's inspired by um in in in uh really interesting ways. Um there are a couple places that we currently use AI.

I you know I think on the on the electro side unfortunately there's not a lot that you can do um in the end there's very tight margin in terms of signal to noise ratio um so there's a lot of innovations that we we've had um in at Neurolink on you know low power and low-noise amplifier designs and being able to digitize that as quickly as possible into digital bits that then you can apply a lot of interesting kind signal processing um as as well as um really at the end of the day this giant pipeline that we call neural decoding you know what is the human intent and how can you actually translate that to something useful.

So uh as of right now we have a you know very simple uh machine learning uh uh you know AI kind of there to um translate those thoughts and intents into something useful and um the the thing that is actually very interesting is that it's not a static data points either right so if you look at a human brain um you know there's what's called neuroplasticity which means that from day to day uh or even depending on the context like there's difference in neural states that are represented even from the same brain region and um it's a learning system, right?

And as well as the the machine learning model, the silicon neural net is also learning about the brain but also your brain is learning about the the silicon uh basically this new mode of communication that you have um gotten as a neural link and uh in many ways I think there's there's some opportunities and we're starting to see kind of interesting traits where now that we have not just one but seven human participants there's some similarities that we see um in in kind of the data set and there's opportunities where we can use that kind of base similarities to improve the calibration time.

So meaning you can't immediately use this device as is. You have to go through a calibration, make sure when you're thinking about moving to the left, we have a stream of neural signals that then say, "Oh, okay. Um, you know, John is thinking about moving to the left and this is what the neural patterns look like.

" And um, you know, collect that data over time and then, you know, improve improve the system. But you do you do get drift over time um due to the fact that as I mentioned your neural uh state is always going through a plasticity. So um can you talk I think as we yeah go ahead.

Yeah can you talk about the the translation and the evolution of like what outputs you're actually trying to map to because if I remember pong pong's unique in that it's it's just up or down. It's not an X and Y grid that you're trying to control. You're just controlling up and down.

But then when I've seen Nolan P1, it's clear that he can use a mouse in an X and Y axis and then also click. And when I think about playing a video game, I'm kind of using all 10 digits. I'm in one way I'm thinking jump. In another way, I'm thinking send the message of just press the A button.

And so there's a there's like how much data can you get out? Are you trying to increase that? Has that already increased?

Is it is it going to uh kind of operate like a higher level of abstraction that then gets translated into computer use or or is it more like as long as I can puppeteer 10 fingers then I like you can translate those into 10 different actions and remap those and have people menu accordingly like the like the investment banker who doesn't use a mouse.

Right. So, um, the the update that we just released, uh, a couple hours ago actually has some videos of, you know, what the latest capabilities with Neurolink, um, you know, that participants been able to do.

I guess one thing that I will highlight is that there's a video of two of our participants playing a first-person shooter game. And if you actually think about it, that's quite sophisticated control. There's a left joystick uh for movement. There's a right joystick for aiming.

And there's a bunch of different buttons for swapping weapons, reload, and shooting, etc. , etc. And so make me cry. It's going to make me cry.

Team death team deathmatch after with your boy after Well, the other thing that's actually kind of interesting is that, you know, at Neuralink, we also talk about, you know, going beyond the limits of biology and then actually achieving superhuman capabilities. Yeah.

So this we're already Oh, I was going to say like we're actually already starting to see some signs of it. Interesting. So that that was my next question. Uh right now you guys are generally focused on outputs, right? Controlling uh an ARM or or playing a video game or or navigating uh some type of computer.

At what point would you focus on or or sp starting to spend more time around sort of the inputs into the brain? Yes. um inputs. So one of the major applications for that is a product that we call blind sight. So giving sight back to people that have lost it and that's primarily going to be uh input.

So uh imagine basically someone who's blind being able to uh have a uh set of glasses with embedded camera that's capturing the scene and that gets converted into set of uh impulses that then stimulate the part of your brain called the visual cortex which is on the back of your head and it basically gives you uh your sight back right because the way vision works is that you have your light hitting your retina converted to electrical signals and at the end of the day it's your brain where you're seeing and having that conscious experience.

But if you have anything break in that circuit, you can directly go to the brain, you know, stimulate those neurons that are giving you that visual experience and be able to see. And our plan is to have our first blind sight patient uh next year.

Can you talk about some of the how you see the ecosystem developing around uh Neurolink long term? I if you can speak to it at all.

just imagining that when when when I when I uh I've I've spoken to Nolan in P1 and uh I mean he's you know incredibly bene he's a huge beneficiary of of Neuralink but also voice interfaces because he can speak and so as that technology gets better that becomes an extra tool in the arsenal.

I'm imagining someone wearing, you know, glasses that act as blind sight, but then also having an audio modulation there that could help as well.

And there's a whole bunch of other and I'm wondering if there's going to be like an ensemble of products or something that where you know, Neuralink's a key key technology, but then there's other approaches that are actually more compatible or or complimentary as opposed to directly competitive. Yeah.

Yeah, I mean I think in the long future, you know, I I I think there's going to be a lot of application layers that are going to be built on top of Neuralink, you know, in a similar ways that when the iPhone was released, I mean, it was it was a cool device, but also it didn't really do a lot until you started developing a bunch of applications on top of it.

So I can imagine a world where there's a neural link app store and in the end it just kind of provides a conduit both into and out of your brain some ways of getting the human intent out to a set of machines and that machine can be your computers laptop or you know your computers phones or prosthetics or some some other thing and and I'm I'm really excited to kind of see what the the the creative minds can kind of think about in terms of the application on top of yeah that it it feels like there will be like a hierarchy of needs within the app store of like, you know, some people might want to use Neuralink so that they can see or talk or, you know, manipulate the world, but then other people might want to use inputs to be like every time I'm about to pick up, you know, a bag of Cheetos just like, you know, don't work, you know, like make the arm not work.

And so I can't I just can't physically pick up the Domino. It's an ad blocker for like overeating, you know, or or turning on the TV. Amazing. Uh I I want to talk about uh hiring. Obviously, this event was was not uh for the shareholders. Uh it was for recruiting. Um can you talk about the Neuralink culture?

Can you talk about what skills you're looking for and what it takes to get a job at Neuralink today?

Yeah, there's um there's one metric that we basically look for in in any candidate and that's um and it's it's hard to actually get it down to a quantitative number, but you know, we try to look for literally ego divided by ability. So, ego to ability ratio. Um and we want that to be significantly less than one.

So, you can have a little bit of ego, but you better have, you know, 10 times, hundred times more abilities to offset that. That's amazing. And uh we we found that to be actually really important thing to look for especially given that one this is an extremely interdisciplinary effort. Mhm.

Um, you know, it's extremely I don't think there's a place where you can have lunch with um, you know, someone who's an expert in RF chip design to robotics to neurosurgery to animal care specialists. I think it's a very very uh unique space in that and not to mention FD lawyers too.

I I I'm always baffled by that that you guys are executing at such a high level regulatory completely different discipline and and I'm saying this to me looking at Neurolink it has to be the hardest problem set in the world. It's hard.

I'm trying to think of others and it's hard to come up with anything that really comes close to it and I imagine that can be really daunting to some people that say I'd rather I'll stay in the Elon orbit, but I'd like to just work on rocket science like, you know, I'd just like to be a rocket scientist.

Um, and so I'm I'm yeah, I'd be interested to get your view on it just because you have so many different factors the the interdisciplinary nature of it.

then okay well you know now the brain's actually moving and it's not moving consistently but it's moving based on you know the way the heart's beating or or things like that.

It just feels like the most difficult problem set and I guess that must be a good filtering function and that you don't want people to come in thinking that you know it's just going to be hard but you know very achievable. You almost want them to think that it's borderline impossible but sort of push through that. Yeah.

Yeah. So you know a lot of people have misconception that Neuralink is a science company. Mhm. We're really a technology and an engineering company.

And you know I there's also this misconception that oh yeah at Neurolink you guys probably spend a lot of time talking about like future of humanity and like uploading consciousness and etc etc. In some ways, yes, like we do talk about those things, but it's like 0. 1% of our conversation. Like 99.

9% of the time it's about hardcore engineering problems and in many ways, you know, brains moving, the material properties, mechanical properties of the brain. Those are engineering constraints, but at the end of the day, it's an engineering problem, right?

So, um, you know, we're really looking for just hardcore engineers. And in many ways, we have a saying that you don't have to be a brain surgeon to work at Neurolink. you actually don't even need to have a prior experience in brain.

Uh I took some neuroscience classes in in grad school, but I didn't really know anything about the brain. Uh it really is just engineering a great system to be able to study and peer into the dynamics of brain that you know we all have that we don't really understand and haven't even scratched the surface of.

So it's a technology first, engineering first company.

for where the company is right now is I I I love that ego toability framework but uh talk about interdisiplinary work is is there is there a world where you need RF engineers who are just great at that and they're not really going to have to cross over or are you still looking for the types that can uh you know bring together different ideas from different disciplines and actually bridge different groups.

How relevant is that? I'm sure that's important at the executive level, but like you might just be in a stage where you know you need to go and solve a key problem in a narrow domain and so you're really looking for just someone who's fantastic at that. Has that evolved over time and and can you speak to that? Yeah.

Yeah. So, um you know, Neurolink is still a very small company. Um you know, we have 300 people uh total across Austin and Fremont. And yeah, there are definitely set of problems where I would say we we need uh GPUs. So people that are very specialized in things or sorry um CPUs.

Um but yeah, like we're still we'll still we're still looking for uh way more GPUs. Yeah, it I mean you're hiring for UIUX here, but then also clinical and surgery and robotics and ASIC. That's fascinating. Firmware, electronics. I'm just looking through it. Machining.

It's really like if you're good at anything, it seems like as long as you're great at it, you can go and work there. Um, I want to I want to dig into uh ego a little bit more uh because I find that very interesting.

Um h how does how does ego manifest in in an employee like like and is that something that can be controlled or or is it is it do you think it's some it's somehow innate or can it be kind of coached out of someone?

When I think of ego, I often think about um someone who's maybe overconfident, but then they might go through an experience that humbles them and then they come out on the other side much lower ego. And so maybe you're not in the position to be the one to take that risk.

Um but but how do you think about the shape of ego as just a human trait? I mean, you're kind of studying the brain. So it's kind of an interesting question from a philosophical perspective as well. I might also consider launching the ego app. Yes.

in which you can, you know, dial it dial it in the app store on the Neuralink app store. Just dial it up and down. Hey, I'm getting I'm getting a I'm I'm going in for a job interview at Nurling. Turn this down all the way. Yeah.

Or if somebody's joining TVPN and we think they're super talented, but the ego's too high, I can say, "Well, you're welcome to join, but we just have to dial it back. You got to install the app. " Yeah. People learn. Uh I I' I've seen this time and time again.

uh especially especially people that are you know uh earlier in their career have um you know a lot of opportunities to kind of shape how they work within an organization and that you know really it's not about pushing for your ideas but it's pushing for what's best for the company and I I've seen that change and um you know I think once you have sort of like it really just fosters like a great environment where people you know debate the merits of ideas rather than you kind of putting sort of their ego in front so that it becomes sometimes like not a technical discussion but a an emotional discussion and um but yeah that's that's certainly coable.

I want to talk about the patients. Uh it was interesting to see how fast you moved on from uh spinal injury to ALS. uh is there a third uh indication that you're interested in in uh in solving in the near term? Uh is this more of a