b0313y This NOTE was important to me because I was speaking with a group of people on Clubhouse about Neuralink and what it really means and people were really receptive, I felt that my take on Neuralink was eye opening to other people based on their feedback and so I intended to use this audio note in particular in the book the copy out my description of the Neuralink, and what it means in the historical sense.
Please note that audio notes were auto transcribed with neural networks via Google Recorder, and then I put them into Google Docs to run an automatic grammar check. So what you are reading is NOT exactly what was originally said. That is why I am sharing it in my notes but I will not share this in this form in my book.
The following is a multi-person conversation without names, I did not write it all, there is an audio recording by the same name b0313y that I will use later to separate out my comments from the rest
(I’m not sure who is speaking first) kind of like a cat scan, you know, like which part of the brain is active and specific functions or specific things you do so going back to the consumer market. Where you know we have like targeted ads now you have something that is you have targeted as towards you and now you have something essentially knowing which part of your brain is reacting when you see specific something specific specific product now you're getting tailored even better tiller internet right now is this kind of getting you to like that lazy mentality of like kind of just sitting there that's my that's my first point second point is going back
02:46
Exactly why, when and what parts of the academic structure that are not interesting enough and now you're essentially putting them all in line. The like even make them into robots in my opinion and like both of these connected this is in to me it's we're getting close to the matrix if anything and like I do agree with you Matt about like, you know like this being the next step but at the same time the next step can also be.
03:12
This you know, like us kind of just being trapped in this mindset where like everything is being watched and not controlled but like targeted that's that's my views, at least. Make the term is the type or you're not a robot. So real quick. I think that's based on the assumption that we continue doing things the way that we have been doing things, which is totally valid but I think there's an equal.
03:39
Potential that with a neural link and with understanding of like look there are better ways to learn than the sit down and and and focus, you know, project-based learning game-based learning there's other ways of learning which are. Burgeoning again project-based and game-based it to go and basically overthrow the you know, the 150 year old industrial kind of way of you know, the ways that traditional education is so if it was that way then yes, I could see it being.
04:13
Versus there's my other ways it might be better ways and I believe and I know there are better ways and I think that neural link would make it clearer that the other ways of learning are better and so not going and continuing this the same way that that sure to make that one point expected right just just to respond to that real quick.
04:32
I do agree but then this also puts the power and the parent set now the parent is able to kind of dictate where and what the child is going to be because they they know how to essentially like use their brain or map out their brain now like that's that's kind of what I'm coming from where it's like, where's the free will of the child kind of trying to decide because now from such a young age the parent is able to control the thought process and manipulate how the child's gonna be.
04:59
Looking for real quick with that with a question so how often have we I know I have daughter said I can't turn my brain off right so you know, so when it's time to go to bed at night and I'm still thinking about things from the day or things that need to be done what implication does something like norlink have you know, with with that the ability to live a holistic well-balanced life and one that's not over-dominated by the thinking compartments of our brain.
05:29
Sorry, can I weigh in here please? I just wanted to say that as far as the neuralink goes, the link is not able to make decisions for you to interfere with your life in any way.
05:54
It's a trip, it's a mathematical solution to something it cannot interfere with, it cannot really feel with your thought process your thought processes your room, actually.
06:15
Because you want to chime in I know we'll go to the insurance but I want to give data a chance to to say to say why? So my thinking is that once you put some obviously like once you put something the electro into your brain like why do you think that it wouldn't interfere with your thinking thoughts like depends on also like depends on what the device do like if it's only taken in information even if there's I mean, even if the metal is up and also like correct me if I'm wrong it I'm putting you out on the stage, but like even if it's a metal like why do you?
07:02
Think that like putting some putting an electro inside your in inside your head interacting with your neuron wouldn't change your behavior that's something that I don't get it the first place because like metal is also like memorizing material chemistry and second thing is that like depends on what the device do if the device is only taking an information then okay, maybe it interacts maybe it would change your maybe 07:37 it would change your structure brain structure it also brain structure is constantly changing by the way.
Also it will also change because you put a battle in because you put electro in the inside your head depends on what obviously depends on what material the electro is, but one thing that also.
07:54
Device in it and showing you like what this is doing then like yeah, maybe it will change your thumb maybe it would change your behavior less but if it's also giving its feedback into your neuron, they're like yeah, definitely it's gonna change your behavior frankly. I think that might not whatever neurons do is just reflect on your behavior and that's like what's up what people have these things about like you when you put.
08:26
In your head, that's one thing and then the second thing is what we're talking about. AI and human well brain & computer symbiosis one thing that we have to think about is I also change your lifestyle as well, for example, let's say you have an implant in your head like the only way to change it only to debug it is to go to the hospital and do it bring surgery like you do you really want that lifestyle like that's something that we can think about right and a third thing is that.
09:03
In here we make an assumption that human intelligence is basically how we stimulate our neurons but like it so like we make because of that assumption we assume that oh having this device in our head we make smarter or you'll make us more capable or digest information better.
09:24
But. I don't think there's evidence for that frankly most of these devices study on. A paralyzed. And like I couldn't see a future like I couldn't see like why are we using it in the normal human being and like creating a new human species? I guess it's something that's pretty sci-fi but at the same time like we don't really know what intelligence is yet like why would we make the assumption that stimulating our Neurons makes us smarter?
10:01
Deep water and in the last thing is about education, frankly, we were talking about it using. Technology to improve education. Especially in China frankly it already happened. There's a company I used to work for that is exactly doing that but it was using an invasive machine interface. Analyzing a student's attention on certain things is like long term attention meaning focus.
10:33
So yeah, I mean, frankly I personally think that like these devices already happened like that's like changing our behavior so like and so far it seems to have an okay result. I think the more important thing that we have to ask is what machine learning algorithms that's using these devices are analyzing and like outputting.
10:59
And affecting our behavior I think sometimes companies put a missionary algorithm on certain devices and claim that they do certain things but then like does it really do that like is it really a valid result like that's up debatable? And but we're using these to change our daily behavior which I think that's something that should be that we should be more alarming about.
11:27
So yeah, sorry, sorry, I took so much time. I'm passing my TV. So I,
Micah Blumberg: I want to tell you a little bit about myself real quick because this topic is what I've devoted my life to since 2012 so I know a vast amount of knowledge about neural link and a lot of other BCI technologies.
11:50
I'm an admin for the Self Aware Networks group on faceobok, and Neurophysics + NerveGear, and other groups related to neuralace / nerve gear / neural link. These are all basically referring to the same idea: the next generation brain interface. That group is really old. It's older than the neural link. I write software if you just look at my bio I brought EEG into VR so we basically have medical imaging technology.
12:31
I brought it into virtual reality and and and there's more but but but I do what I want to share with folks. Is that what I what I know because I basically I study neuroscience pretty much since since 2005 and since there one of the latest things that came out was saying that there's a new hypothesis for short-term memory that short-term memory basically so so let's say the short-term short-term memory is something that you need to remember for maybe 10 seconds, so maybe like a phone number so if I say okay, remember this phone number and then tell it to me in a little bit so I can write it down right that's that's what we're talking about with short-term memory, but You see what a neurons are what yeah working memory exactly yes, yes yeah they refer to the same thing there's many so people have outlined many different types of memory and in terms of you know, but but I think it just boils down to short-term and long-term memory like you can reduce everything either you're talking about short-term or you're talking about long-term, but yeah working memory is another another name for for short-term memory.
13:44
And so the long-term memory involves, you know with what's called LTP or long-term potentiation and you need an any new protein synthesis and a bunch of other steps in order for long-term memory to happen but short-term memory has been a little bit more mysterious and a recent paper came out saying that well so a single neuron couldn't be responsible for long-term memory and this is an old idea because a because the single if you activate a single neuron artificially the the activity is gonna last for I think it would.
14:19
Like 10 milliseconds of hand but it but in any case it's not gonna last for 10 seconds it's it's just way too short for that so in order for an for neurons to remember something for an activity for more than the 10 milliseconds you what what might need to happen is you'd have to have like a bunch of neurons and and they need they need to talk to each other so so one remembers the activity for for for ten milliseconds the next the next one remembers it for ten tell mills admit milliseconds and basically they're if you could if you could have if you have them talk back and forth where they they're basically.
14:54
Then you know the activity is just happening over and over again back and forth. Then they pick the number you could last for ten seconds, but you need basically what's called the neural circuit. I'm like a micro column and and this is um, this is this is probably exactly what's happening so so so if this is true and it probably is true that means that your short term memory is in neural circuits that means your thoughts are in neural circuits and local micro columns a little clusters and this means that a device like nerd link which is designed to.
15:29
Be able to measure a single neuron firing is going to be able to eventually decode your thoughts like literally the thoughts in your head and and because neuralink is capable of closely therapy, which means not only can I read it can write it can stimulate the brain it can change the the patterns of those neural firings, it means it could literally edit your thoughts so we are talking about reading and writing to your short-term memory now this the the the potential for for misuse obviously is is massive and that's it's a big in term for this group, but the bit.
16:04
The potential. For the reason that this technology is almost unstoppable is that the potential for healing there's so many people with traumatic brain injuries this invisible brain injuries like post traumatic stress disorder so many people with with problems like depression and and all kinds of suffering over the entire world and so the so the basically the point is that the the amount of pain in the world is going to result in people being willing to pay for this tech.
16:39
Nology to continue to be developed so they can have relief from their pain is suffering and so this technology is not something anyone can stop coming but obviously the concern is that people don't want to be you know turned into slaves where you have a machine that's basically telling you basically instructing your thoughts basically giving you your next thoughts, can you imagine if you have like the limb machine it's in it's basically telling you what to think and you just think what it tells you to think like that's the capability here now that's the downside the the upside is not only will be able to help people but you could do things like, When you write to the brain you could experience augmented reality and virtual reality experiences through direct brain simulation and you could back up what you're seeing hearing as as pictures and videos that are downloaded for computer if you are if a crime has been committed against you potentially the court could get a, What's it called a a warrant to search oh yeah some sort of warrant to search the the offenders brain to see if there's proof that they actually committed that that crime against you so there's a lot of things to think about but one thing that that is not going to happen is this technology is not going to be stopped because the the people need relief from from their pain.
18:09
Like that's like editing your thoughts. Jesus, are we also going to be able to be like Neo and wake up in like I know kung fu. I'm kind of joking but serious. So this is right now we're talking about editing short-term memory with a neural link, so learning kung fu.
18:31
I guess it might involve long-term memory and that might take up more research and I don't think that's another thing.
18:41
You outline a very measured statement about the pros and cons regarding this but can I ask your personal opinion about what would be the best path forward like what are your suggestions based on your experience so we're talking about a technology that is you need to compare it to owning you need to compare it to basically like brain surgery if you have a neural link device you should have at least the knowledge of of the neurosurgeon before using it because that is the level.
19:16
Of. Of how much you could you could potentially you know screw with your own mind if you if you don't know what you're doing and so basically I mean, I when I see is a future is that basically, you know, the entire world will will become neuroscientists like at people by the time people are in high school, they'll be they'll know all about neuroscience because because you'd have to in this future world they will never be a time in history after this when people are just okay with with this technology because the risks never go away that what I mean imagine someone just all it takes is like one.
19:51
Person to start turning to secretly turn another person and do it into like a slave or a clone of themselves and then another person and then another person and like secretly just build like an army of like of slaves that take over the world little by little if that is always going to be a risk for the rest of time.
20:12
Yeah, go ahead.
20:21
With our iphone so we're all on the apple platform, right? I mean, I'm sure we've all found ourselves in the eye, you know the Apple store because you know, our battery was dying too quickly or for some reason so so with something like this how do we know if it's broken and then how is it fixed and what are the implications, you know might have bring on early onset dementia have you studied any of those implications or potential implications?
20:47
So to so what what my I study these topics my when I when I'm thinking at the moment is that dementia Alzheimer's Parkinson's that that they're in in these diseases you're seeing basically a new research basically from from 2020 and 21 is saying that there's a problem with new protein synthesis in both Parkinson's disease and an Alzheimer's and if you give people, Chemicals, if you give animals chemicals that improve new protein synthesis, then their memory and they're in their in their functions improve but but what that what that could point to is that?
21:41
Basically I mean but it boils down to it looks like those diseases might be caused by sugar but more than anything else that people are eating sugar but I don't think that so cancer is a specific question. I don't think that neuralink as a device would be a cause of Alzheimer's or Parkinson's. That's my short answer.
22:01
I'm sure this trigger in this group would love to hear that.
22:09
Sorry.
22:15
Okay, so there's there's a I think there's two I think there's really two quite you've kind of one good question but I think there's really two good questions in there that I kind of want to address so one is um, The in terms of like ethics in terms of the laws the ethical laws that need to be applied here.
22:37
I mean, it comes it comes down to like this is a this is like, you know, it's like owning a gun sort of it's like owning but it's like also like being like I mentioned like being a surgeon like you'd have to have the knowledge of what you're doing and you what you have with the neural length devices a very is potentially dangerous in terms of like, you know, even if even if no one has given you a malicious software if there's a bug in the software there.
23:05
There's potentially very serious problems that that could happen and so there needs to be of course and so in terms of safeguards software safeguards there would have to be you know, a lot of there'd have to be a higher standard of of bug testing before software goes into something that you put in a human being there would have to be you know, like the um, you know, you need to be able to verify that the software is 100% what you know, what did it what it is?
23:40
I mean that you'd have to you I don't think that like right now like the current the current thing is that you know, there's there's a Bluetooth connection to a phone and that's just right for hacking right is there the million ways to hack bluetooth there's a million ways to hack your Android phone that's too risky for to put into a human being but um, I think that you know, so the devices need to be more secure and then yeah, so it just has to be regarded correctly, it has to be understood that this is a Super this is a potentially dangerous technology that has enormous potential therapeutic and medical benefits so you know, it's like doctors can get you know, for example doctors can prescribe extremely dangerous drugs in certain conditions for example and doctors can perform extremely dangerous surgeries but not everyone can and not and in that so those this is the kind of thinking that needs to be done about this, um technology before it before it goes out more widely into society.
24:46
I mean, it's like you could potentially turn somebody into a human robot like the way you describe yeah the yeah, it's it's it's both input and output that could be and right now. I know you're linking only works in one direction but yeah like the fact that you described how you're working memory like your consciousness your what you think you're thinking and that you have free will could potentially be intercepted in.
25:17
Bicycle by an outside force that's almost scary yeah well it's not that we want to stop that from happening completely like you may want to emit you may want to enjoy virtual reality via direct brain stimulation you may want to enjoy augmented reality via direct bearing stimulation so you so you could walk down this the street and you could interface with your memories basically is you know, like virtual reality programs that you can see in here and in maybe that the the benefit to that will be profound and enormous you just what people are going to be worried about.
25:51
Of course if. Is that you know that a program could glitch in cause harm if there's a program error and and people will be worried about of course malicious malicious programs, but but the the same source of there is a there's an existing sort of legal structure for you know for for dealing with criminal activity right and and so what would we need to happen is is people would need to think about common sense laws for for.
26:27
For making this technology something that's safe and something that is whether there's a pause right there because I'm yeah. This is if you want to share your thoughts. I know you are needed earlier, um, yeah well first of all, I just changed myself to be I work.
26:55
Between 2002 and 2007 yeah implantable intracortical. Even ushered in what's going on but we can't hear you yeah.
27:14
If we can't hear you right that's probably better, um, yeah those things got just briefly introducing myself. I worked on a project called brain gate which was like the precursor to the neural link those an implantable intracortical motor neuropathic device, so we kind of built that between the 2002 and 2007 around and we actually it was originally a monkey BCI that we allowed enabled monkeys to control computers directly using thought and and then transferred it to human clinical trials, so we successfully in.
27:49
Enabled people with like spinal cord injury and quadriplegia to control computers and robotic arms and stuff like that so it was actually the basically neural length is actually just replicating what we did and what other people have done since then and they're the main major major difference which is very significant is that they're making it as a good commercial project like that is going to be similar to like a LASIK eye surgery type of implant where's ours was like more similar to like a monkey neuroscience research laboratory equipment with like a cart with like all these big machines that do decoding and signal to processing.
28:24
Not very portable at all and their neural link is doing all of that basically on a little trip inside your head which is very impressive and but yeah, I'm just gonna comment on one thing about the ethical guidelines, you know, if you're talking about a computer I think a big difference is the fact that you know, if you're gonna do like a neural augmentation type device and it's implanted and you don't you learn no longer have the option of that.
28:53
Not using it for example so like you know that might be the best ethical type of thing again on off a physical on/off switch which I'm very very unlikely what they have you they tend not to want to let that go but for example with it iphone or a computer you can just turn it off and just leave it at home or something with the neural link, you will not be able to do that and I think that's where you really need to have ethical guidelines will aren't where it's not just like faceobok or Gmail that reads all of your direct messages and personal documents and things like that, there will have to be some kind of a little bit more stringent.
29:26
I mean, I I shouldn't say there won't. Have to be I'll say in an ideal world there would be more stringent ethical and privacy guidelines in the more most likelihood like there's won't be and we're just gonna have to you know, whatever and like at the the extreme cases that like what Mike was saying when there is like a subpoena or a criminal investigation, they're just gonna be able to you know, it's a it's it's gonna be a US company the government will just be able to subpoena whatever they want.
29:53
I also just wanted a differentiate between. The kind of short-term and medium term prospects versus like longer terms like anything about like thoughts and decoding thoughts and stuff like that that's like more like medium to long term in the short term is going to be basically moving a cursor on the screen maybe typing things out very rare memory typing and things like that and it's just important to know like that that is where they're at now and probably will go and they let next to like maybe auditory auditory and neural press these days or visual prosthesis and things like that and maybe some treatments of like some motor neurode motor disease like Parkinson's or You know those kinds of things and then yeah, I think it is definitely true that there will be things related to thoughts and memory and that will probably come in like the 10 to 15 year timer as or maybe maybe even like 7 to 10 year but those aren't coming like right around the corner and I can I can speak to like actual to have those things to work in the motor neuroscience and decoding from populations of neurons in the motor cortex, which is pretty interesting stuff but.
30:59
Yeah, so that's it you touched on some really interesting areas and would love to dive into one one in particular was around like like the regulation like the way I you know, like from hearing Elon Musk talk about this, you know, let me use the analogy of like let's say we alien space invasion or like the movie independence day if we knew that there was going to be like this superior species that was gonna invade earth or visit planet earth and we had to like it advance warning of it to.
31:34
20-30 years in advance. You know, we would probably rush to prepare for that situation maybe and my guess is that he he's viewing AI as essentially that and that regulation seems to be always I think that comes after something catastrophic happens and and the damage is done and then we pursue regulation and that idea of pursuing regulation after the innovation as you know played itself out a bit to see it's the edges of where it might want to be might want to safeguard the, Is.
32:12
In tech that process of regulation being a laggard is a bad one, so in essentially the only way to combat this invasion of AI is us also kind of brushing on the speeding train towards becoming a cyborg and that's essentially you know, the only thing that he sees as a passport and I that's kind of what I'm hearing.
32:36
I don't know I'd love to hear your thoughts and see if that's what you're essentially saying as well, can I add can I add something. I don't want to reply. I just answered Travis's question first and so I think I will travel. Well I think the first thing I would respond to that is like there's a lot of hype around AI different things like that and you know aliens and reactions like things like this like I don't know I use important just look at like what we're talking about is like decoding the let's say motor intention from the from the brain and whether using AI or using like different types of like signal processing or real-time there, they're all kinds of algorithms and different control system types of decoding mathematical ways of doing that it's not just AI and these I think it's important to just like, Just kind of like look at all these different aspects of like neural interfaces I I guess I'm sorry so yeah so I was personally I think we're regarding like the ethical guidelines I hurt like I think it's also important to differentiate those like just because you're doing a neural interface doesn't it's not that different.
33:43
I think that's one thing I really agree with what you want when Elon Musk talks about this and he says that you know, this is like we have in neural interface right now, you know, we have we have a keyboard we have a phone we have a mobile phone with direct access to the internet with.
33:57
Like hundreds of gigs of memory and and pictures and facial identification and Wikipedia and email that that's a neural interface you know you're using your thumbs and you're connecting it to a mega database of like real-time AI and all kinds of like signal processing and everything at your fingertips, it's just a very slow bandwidth between your thoughts and then the computer systems that are connected to the internet that can pretty much connect any other computer system all around the world faster than your eyes can register it in the microsecond millisecond range, so we already we're I'm talking to you right with a neural interface.
34:33
That is sending my voice to your ears it's not really neural direct neural interface but like so it's just a very slow bandwidth and I think my point there is the fact that we don't need to wait until we have a neural interface to have better ethical guidelines we in my opinion like why don't we start with the current interface which is like using a computer like right now we're using this app and we're using an IO iPhone and we don't have any control over the privacy or like ethical guidelines on those systems, why don't we just try to get those regulatory because in my opinion I'm saying that we should do that.
35:05
I don't have a lot of faith in the car. Companies and in the government that they're going to do that and then the issue that I'm concerned about is like now when you have an implant. Now you don't have the chance to turn it off you can tell it's just implanted and now they they just have you in your own brain and it's kind of like if we should invite you that should be like a big flag or a milestone which like let's put the brakes on it now but we should also work backwards into faceobok into Gmail and things like that may I was mayor respond to apes today sure okay so soaps work is extremely extremely legit like what abe has done in terms of intracortical implants.
35:50
E-cig and I don't know if you're also working with stereo. EEG or DBS a. We might as it was an intracortical one it was not eco it's a intricate like eco is on the surface of the of the brain and intracortical is the actual implanted with like listening to the actual neurons firing and firing rates of individual the populations of individual neurons and just by the way people know it it was a nature paper.
36:15
I was a co-author on a nature paper and my claim to fame as I was actually on the cover of nature and July of 2013, well what was the name of the paper so we can all look it up and we're working with DBS. I think my divorce is just being signed up there now.
36:34
Okay were you working with DBS and also can you say the name of the nature of the nature paper we can all look it up, um, if you just go to nature.com and you go to like or yes neural a neural prosthetic and you can is on your hockbird if you look up h o c h, b e r, g.
36:50
h o c. H, b r. h, o c a h, oberg and just look up neural neural prosthetics. Okay, you see he was called neuro in some ensemble control of prosthetic devices yeah neural yeah you're just like Google neuronal yeah neuronal prosthetic heart vertical it'll come up h o c h b e r g.
37:15
Was it a 2006 paper that's the 2006 yeah there was another one 2002 which was the monkey controlling the computer and then in 2006 it was the human controlling and yeah, you can look up that whole addition of nature in July 13th 2006 has a whole set of articles and editorials.
37:37
This was an actual published article and then there was like editorials and other documents about neural processes, it's pretty interesting and yeah, we didn't I did work with DBS in the sense that like our preliminary test we did before we had the FDA IDE clinical approval to do the device testing in in our own trial we actually pay back on a implanting a DBS there's a there's a technique where in order to target the exact location of the DBS electrodes, you can use electrodes that they normally put in deep in the brain to find the exact spot.
38:12
Where you want the DBS implanted so we stopped those electrodes on the surface of the brain this made by coming called alpha omega as an Israeli company and we actually modified their system to do a real-time decoding and control allowing patients in the OR while they had their brain open for the DVS implantation we got the patient's consent and IRB approval to do a closed loop real-time system in the OR so I was working with patients in in the or getting a DPS implant so but I didn't work on DBS, but we used that OR.
38:45
Time. So how close were you able to get to measuring like single neuron spikes? And I asked that because I guess one of the major differences that people have been saying about neural links is that when you measure single neuron spikes you're talking about a new order of magnitude of resolution potentially for medical imaging.
39:12
It just doesn't exist and in many other devices and any other device that does the whole technical that's what, I mean, that's it was. It's pretty standard in monkeys. In the mid-90s or early 90s we were doing it in humans. In 2000 I think four or five and it was basically spikes.
39:32
So we're recording from between you know, five to a couple hundred neurons in parallel at the same time and you're getting the full spike trains with fine timing the exact, you know, microsecond spike timing. So, okay, so you're talking about an electorate race that was implanted in monkeys, for example.
39:51
In humans in human electrode arrays in the human motor cortex and decoding the spiking activity of populations of neurons in real time. Yeah, okay. And so do you feel that we're getting anything new in terms of the resolution and capabilities with neural links? I am just interested to hear your thoughts on that.
40:12
Don't definitely, I mean, Generations.
40:20
Their first experiment is is pretty they're setting a very low threshold and and very easy thing to succeed because it's been done it was done in 2004, okay like this is this was this is not a complicated thing in terms of the technology in terms of replicating it because all they're doing is doing like a kind of control of the iPhone or a computer and there are technologies way more sophisticated and my understanding is there's like thousands I believe in in their current version one is gonna be like thousand plus real-time decoding.
40:52
Yeah, I think I'll time out. I think the current number is over three thousand. But the device is potentially capable of over more than ten thousand. Yeah ours was maximum it's called the Utah rate that's like the classic array that name is called the Utah Ray it's a hundred and neuron sorry hundred electrodes on a silicon it's like a it.
41:12
It's actually like a 96 volt 90 says a hundred electrodes on a silicon wafer, so there's like a 10 by 10 array. It's just a four millimeters deep in the cortex and you just tap it right onto the surface of the cortex and just the tips of them as are coded in metal that you can so that has like, A ability to record the electrical activity very tiny electrode that can just pick up just the field around the tip, which is basically picking up the actual neuronal pattern near the neurons that are nearby the tips of those electrodes.
41:47
But when you stick a Utah array into a human brain, I mean, it looks like a it has it looks like it like a square with a bunch of spikes or are we we're puncturing a lot of blood vessels right causing a lot of damage and then scarring happens then this helps working is it right?
42:00
It lasted I think the maximum was like a couple years and there there's not a lot of scarring, I mean that there's a tiny amount of you know, it takes the thing like a week or two to heal after the brain surgery which is really common and it's just like the first time plug it in is starts working yeah but yeah, definitely it's the the neural link is less invasive because they actually use these tiny wires and they just thread it in very precisely this this thing is like a little more invasive but um, you just avoid any major capillaries and arteries, so where were you were able?
42:35
To send a stimulation on and do close through closed loop therapies we weren't doing simulation but you can the neural effects originally developed to do cortical stimulation individual cortex and they did it and they you get like phosphines and things like that it didn't that you can stimulate with the Utah array we didn't in our system we only recorded from the primary motor cortex, okay, can you be very specific about which which pin the the simulation is coming from in the entire right or is it just like the whole thing at once all of any of them you can just send current into any of them?
43:10
It's just it's just you could be it's just an electrode so you can record or stimulate very very cool so um, one of the things we're getting maybe very specific. I am looking at seeing some attrition like this so maybe we want to get more general. I don't know if people are interested in this level of detail.
43:23
I mean, I like talking about it but I just thought okay good question this why don't you go ahead sure if you have good questions were you in ducks whoever laughed by any chance? I know oh okay that was in the dawn of you let Donnie you love at Brown and then we and that was they were the ones who were I mean Don he was like quite amazing prominent motor neuroscience researcher from back in like the 80s and 70s and he was basically doing monkey, you know, motorcycles motor neuroscience research for a long time and I'm they yeah, that's where they did the initial monkey closed loop intracortical that system and then he started a company called cyber kinetics, so I joined that company.
44:08
I was kind of like the first employee hired and then we merged. With a company in Utah that developed Utah array and also the amplifiers and decoding systems, they still exist and sell those systems for neurosurgeons research as well as clinical epilepsy research and things like that.
44:26
Yeah, I know I'm . I think like some of the labs that I participated in actually used. Utah array there are some electrode failures and I think I also pointed out that there's also a risk of a dema when you're puncturing a blood vessel but I think they're kind of widely used for ring computers and faces.
44:50
A lot of people have used them now and it's been widely replicated In fact they're people who have used them like they implant multiple. Utah arrays and different parts of the brain and then they decode. They use it like they have demonstrated it to be used for hand and arm robotic prosthesis.
45:07
So someone who loses their arm like a veteran had their arm, you know agitated and then they put a very highly dexterous robotic arm. You can see these demos on YouTube and they say that they don't usually use the frame computer interface for those but they have demonstrated the ability.
45:25
To use a brain interface to control fingers and hands of another complete robot and hand-in-arm with like 25-27 degrees of freedom. So you guys have a really interesting conversation. I just want to remind you that there's a lot of books that don't have I guess some of the technical details that you guys are referencing.
45:45
Could you perhaps just for everyone in the audience explain in layman's terms what is the Utah array and how they kind of get the sense that it's different from what your link is using? Could a baby maybe explain that a little bit more. I would love to hear.
46:00
Understand that. Um, here I'll appear. I yeah, basically Utah Ray is like a kind of vegan thing of it as a traditional intracortical electrode like you can see.
46:22
Sorry. So PTR means uh went once abe uploads his picture, you're gonna pull down the screen refresh hit to see his new picture. Yeah, basically Utah rate and this is actually from the intro paper. So yeah, so yeah. Just like.
46:45
So that's what it looks like. It's basically a well just one more second sorry it's uploading I guess yeah it's it's basically you can PTR now and then you'll see that Utah, right? It's basically like you can think of it. I mean for lack of a better term a bed of nails.
46:59
So speak it's basically a silicon wafer four millimeters by four millimeters and one millimeter in in height and then it's etched out so there are a hundred electrodes ten by ten array and I don't know if people are familiar with like semiconductor etching but it's like a mems device so it's kind of like the silicon wave for like a computer chip and then It's basically every electrodes just like a millimeter sorry nicely.
47:25
Yeah, basically it's four millimeter by four millimeter in the square and then the electrodes are one millimeter in height so that the electrode goes one millimeter deep into the into the cortex which is just the the act that correct and required depth to reach the neurons and if anyone is more familiar than I am with the modern neuroscience, but there's a specific neuronal cells in the inner motor cortex and the cell.
47:50
The actual cell body is right around that layer. So we're actually recording from the ideal depth in that using that system and it basically it's very different from neural link in the sense that it's a rigid it's like one flat surface that the electro tips are all at the same depth and it's just what happens like work pretty well and but you don't have any variability.
48:15
You can't just place the electrodes in particular spots. So you just kind of like put it there and then it works in all the patients so that was good. But it's if you can think of it as a rudimentary system that was more developed for monkey neuroscience research.
48:31
Okay, maybe I can create a more like a story just what it does. You neurons can reach this one chip at this moment. What's the maximum amount of neurons these trips can I pay in theory? I would activate I mean in theory every electrode can actually record from more than one neuron because if you have an electro tip and then there can be multiple neurons around the electrode and then you basically see different like every electrode every neuron has like a different different pattern of the action potential so you can actually see multiple neurons around the tip of an electrode.
49:11
So in theory, you can also, Like 200 to 300 neurons but in practice it would typically range due to like possible scarring or possible, you know, not proximity to a neuron or something like that or just a broken electrode that can happen. I'm likely but back when we did it in the first first times was like 2004 or 2005 that is important to note.
49:31
I'm talking about some pretty old thing that was like 15 years ago. But back in those times, it was probably more like between 20 to like a hundred so that was pretty and what is it currently right now? Do you know what the current amount is like?
49:46
We it should be around thousand hundred and fifty I would probably ask oh yeah we haven't reached a thousand yet Well no the neural link will for the neural link will be different in the in neural link, they're actually going to be inserting wires like a tiny filament wires and targeting everyone will be implanted by a robot that's targeting specific neurons and they can implant like a couple thousand probably three thousand into the into this specific area in the brain individually and then I think they can actually record from multiple neurons for each wider, you can actually record multiple neuronal like recording sites and stimulate.
50:21
So it's like is like orders of magnitude more neurons and you're able to place them precisely so that the Utah array was like I said like a flat sheet of like a hundred recording electrodes and you get whatever you get when you implant it where's the neural link is like precisely implanting three thousand individually so that's it and two clarified.
50:44
I think the best metaphor that actually worked for me when I was, you know, working, is like you know volunteering some of the lapses that it's like dropping your mic and in the middle of the stadium, so These. Electrodes are doing is kind of like listening in to different neurons talking but there are like I guess like one of the novelty novelties have neuro-like is that they're able to do spike sorting or mechanism that's actually differentiates like which neuron is firing it would which location then you can you can essentially decode the neural information is to like which neuron is like firing it what rate so it's like like I said to go back to my analogy of dropping Mike and in the middle of a stadium is Like a you're trying to discern with which person is talking based on their voices but the only instrument that you have is the mic so what these electrodes are doing even though they're not reporting like for a signal neuron but YouTube different clustering type aren't like by applying different clustering techniques, you can actually decode which how many neurons you have in the the constellation where they see selectors are implanted.
51:59
Well you get a lot of information actually because we're in a specific area designated for like hand and arm movement of the brain of the direct in the primary motor cortex of the brain that controls the hand and arm and it's it's like the last part of the motor system that right before it gets the signal gets sent down the spinal cord to the to the actual hand in the arms, so it's actually like very rich information and even if you just have like five to ten neurons you can actually decode with relatively high precision the exact either location or velocity of the hand in space and that was you know, so if It's surprisingly how precise exercise surprising how precise enrich the information even from the relatively low number of neurons and and once you get more information and more neurons you you just you don't need it it's it's yeah the amount of encoding and the is is quite impressive in the brain and yeah, I mean if we want to kind of generalize this more for the yeah, I think we've gone a little bit to in depth on these specific questions maybe but maybe if anyone has other general questions or topics to discuss yeah, yeah, I just Elon when he first, Really first presented neural link the the referenced this an application that you know, I think either was FBA proved or had gone through that process.
53:22
I'm starting to wonder whether he was referencing your work or if there's a number of don't know that he was talking about our project he tied he named it by name at the first presentation done a few presentations in the first time he's he's actually talked about brain gate and all that the second or third time he I think he started to degrade the description and start calling it like a, An archaic system an ancient archaic system that did but I mean in reality it's it's literally they're they're basically replicating exactly the experiment that we did and I think that was done on purpose because of the FDA it in FDA world.
53:58
Things are very slow and they like to see proof and and they like to see precedent and things like that so I think they're probably modeling it after our system for that reason that we did a successful ID this is awesome because this is this cuz we're I'd like to probably take the conversation a little bit but we were just talking about this earlier today on a group chat and how like neuralink isn't is more of an innovation on a prior invention and that he had referenced this thing earlier and then as soon as you started talking it was like could it be him could he be the the could this be the original thing that he was referenced?
54:33
To him, this is crazy that we have here on stage. Can yeah sorry also I wasn't. I wasn't planning to. I didn't know this room existed and I've been having a really tiring week and day so I'm sorry. I'm a little tired else live that's all good no this is this is exciting it's accelerating to have what is essentially like a key person in the founding team of the thing that really I think from from the perspective of an innovator from the perspective of what Elon I think it was the work you guys did and that the FDA component of it that you had advanced it so far along on the Regulation side that caught his attention that there this was a promising solution and so it's really incredible, you know, the kind of influence that you've had in what is potentially the future of computing and we have you here on stage you reference, you know, the commercialization aspect of what he's doing maybe you can touch on a little bit about you know, has has the prior work has there been prior attempts to commercialize, you know, the history behind this in and and you know, what is it that he's doing?
55:43
What right as it relates to the commercialization of this stuff it'd be really cares and then also just your timelines you kind of touched on you know, two to three years maybe typing then auditory 17 years you said maybe you know that thing that Mika touched on the the working memory being edited or that sort of potential could you maybe share some of that as well a little more of your history and and around, you know prior to mirror Lincoln what you think they're doing right in terms of commercialization and then your timelines, okay?
56:14
Travis can add something well, can I may I add something. Yeah go ahead so what I wanted to say is that Abe is an amazing guest to have but nerling builds on I think a hundred years of neuroscience. I think you know, like what nearly builds on not just apes work, but also, you know neural the work on neural dust and and optogenetics and it's is it I believe it represents a true successor to to EEG and eCog and DBS it there's there's tons more it's like it is the next frontier of 100 years of neuroscience, it's yeah apes work is is really a big part of that, but This is so much more it also had like I was like employee and I was just out of college and at that company so I was the it was the Donnie lab in awkward doctor professor Hogburg and these other guy mehale Misha Ceruya was actually kind of like the the star guy who did the he was his PhD thesis was the monkey the closed loop monkey intro cortical monkey, you know motor like basically a neural prosthetic in monkeys, they they got that was kind.
57:28
Of like, The major the major milestone in around 2001 2002 and yeah those guys kind of made it all happen. I kind of I'm good at just getting things done so I kind of joined and I was coming from a computer engineering background but so we yeah, but that's guide you right now.
57:47
I'm glad you reference that thank you Mica. Maybe like you, you're part of that larger team and the larger movement that's been going on for some time, so I'm glad for your emphasis. I do love to hear you know, like has there been other attempts to commercialize and what are these guys doing right and what do you think?
58:03
Happened in the next two or three years. I mean, I think the big thing about yoga and that's absolutely I mean, the one thing is that it is an intra-cortical decoding, you know, motor neural prosthetic device that brain gate was one of the you know, first systems of that type with populations of neurons we decoded in real time, you know, EEG, it's a just a little bit different technology as well as ECOG.
58:32
I mean, of course all of them neuroscience isn't necessary and like it was just a massive field that. Goes back hundred two hundred years, so yeah, so I mean, I agree it's all it's all part of the same development there but as far as like the commercialization I think it's important to note yeah like this is a huge the neural link if they pull it off which is definitely hasn't happened yet, you know, like they have to have human, you know subjects and and have people controlling the computers.
59:04
I don't. I have no doubt that they're going to do it but, If they get it they're able to do that then. Yeah I even get this is hard to describe how big I mean it's kind of like similar to the scale of like going to space like creating a commercial space program the first in human history, right where before that it was either the US government or the Russia the Soviet Union or China doing that and then it was like a commercial company sending satellites to space for voting the international space station or sending a man missing mission to Mars which is probably going to happen also so it's like I think it's a long those lines it's it's a pretty huge development because commercializing something of this type.
59:47
And mass-producing it if they actually end up doing that which is what they plan to do it's that's it's just it's not an easy thing it's technologically it's not it's not complicated like this technology has existed we it hundred percent in fact it existed before 2003 or four when we did it it was not new technology even then is a little bit complicated to take, you know, neuroscience technology electrical engineering technology clinical neurosurgeon types of things and put them together but it's not none of those things were new it's and they're definitely not new now but putting.
01:00:22
Them together and making them commercial and using like a robotic implant and mass-producing it that's that's very new and I don't think anyone has ever done a neural, you know, a neural interface and I think the coolest part is going to be when they go into an augmentative type of approach going instead of just like helping someone with spinal cord injury or ALS to regain, you know, limb movement or control a robotic arm or control computer or an iPhone if they start going in.
01:00:52
I've mannative that's even a bigger thing but I guess to answer your question. I mean, there have been a lot of commercial projects, most of them on a much smaller scale. I think of anything, I mean Mike, maybe you didn't know something but like on this scale.
01:01:08
I mean, it's the one thing that for me and my I mean, this is not gonna sound like it's gonna sound like in a different realm but in my opinion the most similar thing that's happening now is control lab CTRL labs, which is an EMG decoding device that decodes hand motion.
01:01:27
And you just it's a wearable EMG system and it was just acquired by faceobok for I think over 500 million dollars and it's basically able to decode like you can type without a keyboard and you you can even when you get good at it you can type without moving your fingers at all so it's basically a very high bandwidth interface and it's the guy who created it his vision was like, Doing neural interfaces and neural technology so like neuroscience type technology, but why stop at like medical treatment why why why do we put billions of dollars into like computer software and like iPhones and like VR but then and we have all this neural interface technology, but we only do it for like treating a disease like why do we stop there so his ideas let's take that technology and then go towards increasing the bandwidth between the human and the and the computer system and that's what they're that's what they're doing with a control labs in my opinion.
01:02:25
That's the big things like augmentative systems and then and then there's a whole other realm physically so may I maybe I'll add something valuable here is and that is a control labs it in terms of like the application of brain gate in terms of giving people who who can no longer move their limbs control labs may be a very valuable tool and that that specific context and and it does make sense to compare control labs to neuralink and only that respect in terms of controlling devices, but but um, and and that that that and I say that because the the, Difference is between what the machines are doing is vastly different as a nose it's not news to eight but um devices that might be the might be more similar.
01:03:09
I mean, you can look up the yeah and it won't be that useful for you to look up but you can look up open water by the Mary Lou Jefferson look for her talks on YouTube to read about what she's doing. I'm sorry to watch her talks on YouTube to see her talk about what she's doing . You can look up a few rocks microwave imaging there. Unfortunately there's not a lot for an average folks to read on that topic and I'll try to write something about it.
01:03:35
There's a so that there's more stuff out there there's also there's there's work and you won't be you're not gonna be able to find a whole lot of information about new work on combining deep learning with with electrical impedance tomography that they're seeing some huge medical imaging results therefore for especially for imaging brain imaging the brain that there there the the idea that we can do so MRIs are too slow they measure blood flow with a measured blood flow really slowly so the idea that we could we could speed that up with.
01:04:10
With. New with new faster chips to basically detect neuron firing basically by reduced by by reduced by reversing the refraction of light which is called holography and doing it fast enough so that we can notice the expansion and contraction of neurons as they're firing without enable us to basically read things as as fast as an early link is reading them as precisely as nervous as an early because reading them, but in a non-invasive way and and with ultrasound we could stimulate the brain also non evasively so this so this.
01:04:45
This technology that neuralink is doing potentially could be a non-surgical non-invasive and in in in in the future and in that timeline, I don't know what it is, but that that it also be like a 15-year timeline but the the idea is that maybe you don't get brain surgery and you just sit in the wrong car and behind the headrest there's a there's a there's there's like a there's a there's a device that will wirelessly take over your mind so so that's another possibility and if in the future world but I want to break down also so for for my layman's perspective that the, First vast differences in in what neural link is in terms of we talked about how many sensors and how many electrodes that maybe the deferred the first devices we're talking like 3,000 electrodes and you know, each hair is like I think 16 electrodes and then you have to have a whole bunch of hairs to get to 3,000 and then you know potentially devices like 10,000 micros and each electrode is potentially reading from you know, several hundred neurons and when you have many of them in a string you can do you can do things like you can you can apply you can basically create a really precise maybe 3d map a spatial map.
01:05:55
Of of around the electrodes with firing and when but this is like compared to the Utah array compared to DBS compared to any sort of electrical reading, you know done since you know electors have been implanted in in people in an animals, you know, good for for for very long time over a hundred years.
01:06:12
I think that what you're getting is not only like a massive massively greater amount of information at a massively higher sampling rate and just like you're getting big data, you're getting data that might be more than your phone can handle and your phone may not be able to handle.
01:06:30
The the effect I don't think that your phone's gonna be able to handle the actual like sampling rate of of EEG currently like but in this is vastly more like this is like way more than you know, I mean so so with this big data, like doing things like brain gate, this is um, this is stuff that is way below the capability of neural link in the long term the the real capability of an early I believe is actually in in decoding the the the signals that are passing between the micro.
01:07:05
Clusters or neural columns or neural circuits and and really really beginning to decode what your thoughts are and then beginning to to manipulate your thoughts, that's that's that's the bigger longer term picture as apes. I don't expect this to happen in the first year or anything and I don't know if the timeline is 15 years but I think this is real.
01:07:28
For people in the room and make I got two two quote follow up questions for you here but for people in the room we were listening in like we're the topic is of course neuro-lincoln and whether it's revolutionary or potentially has tragical consequences and there's been discussion on both but you know, we're very fortunate to have some folks here obviously are at the front line of some of those rep this work and long before your like showed up so the two questions I had for you guys is you know, you referenced some of this prior work.
01:08:01
I'm curious. About where where the scent like the geographical locations of this work is there are you seeing you know it you know prior to near Lincoln even now I'm sure there's more companies spending time in the space are there certain cities certain areas and I'm curious but for two reasons one regulatory reasons and I'm wondering whether you know, certain countries have perhaps different moral facilities philosophical differences, that might be okay with certain types of research and so that's the first question you're.
01:08:38
Gonna rapidly speak where all this research and work is happening and then the second question I can do is follow up so I get the answer. I taught my son who is also very familiar with a lot of these topics in general and also Nick and and and a couple other people so I just want to make sure that they had a chance to share their thoughts because Mike and I have both been going babbling on for quite a while and I thanks again for the help for pointing it out pointing it up.
01:09:08
Oh that's great yeah anyone anyone of you guys you could just share some thoughts on geography where's all this happening. I mean with computing obviously the valley played a huge role in pushing things. Is this predominantly in the valley as well or are there other cities that are leading the charge?
01:09:29
Yeah I'm not sure I can speak to just the basics like country-wise. I mean, most of this development is happening in the US a big cool company that comes to mind is neuro paste, they were quite one of the innovators for a long time doing epilepsy treatments, they had closed loop epilepsy treatment, of course like metronic at the implantable DBS there were several other companies that had I mean, I for me it's important to differentiate between kind of reality and you know where in reality as far as like treatment goes.
01:10:04
Not reality but let's say clinical treatment or clinical implant of like patients even in a clinical study or commercialization is like is is one area of something so like neuralink actually had a clinical trial successful clinical trial and then actually had a commercial product that they were implanting in patients is in a clinic in addition to like Medtronic with the DVS and I think that is significant, you know, because it's you can do a lot of stuff in the lab even or just talk about it, but that's not the same as as as having it done and then actually, Actually about like other two graphical locations like that's a there's something called medical tourism and you basically you you actually go to a different country to get medical treatments and it's cheaper and that's usually the reason people do it and you get very high quality medical treatments for a lot less money and insurance companies even cover that now it is because it's cheaper with the flights and the hotels than it would be to do the same thing in the US and there's other reasons to do medical tourism which could in the future be like you go to like an island or something or a place without a jurisdiction or just a country that has red lacks laws and Then on that topic and then you have a very high quality system like neural link or something and you get implanted there and there's not as many restrictions and then you can get that for like an augmentative purpose or something like that.
01:11:25
I think in theory it's possible because I guess it's very likely that the FDA is just not going to open up applications like augmentative types of things or treatments, if you don't have a spinal cord injury or Parkinson's or ALS. Can I ask you that question well, sorry dad.
01:11:43
It was just a five second other thing that French is also catching up with you so yeah which country which contributed right okay there's also a Brazil nicoleta in Brazil geological work in Brazil raised at times. I had a quick question just to sort of follow up regarding the geographic region.
01:12:05
I think Travis posed a really interesting thought. Some you know, there are some countries who have proven that they will use it. You know technology to sort of to control the population or limit on the the freedom of the population to some degree, what do you think based on some of the issues that we're brought about by Mika and and during those conversation are there any thought as to the threat against not necessarily I guess other people but inside the country itself to utilize this technology as a sort of control method or you know, like a hype like hypothetical would be forced implants where your thoughts would.
01:12:49
Be decoded and if you you know, if you attended a meeting or viewed something that was considered forbidden speech, you could be essentially prosecuted even though because they have direct actors access to your thoughts in that regard yeah just thought of like I'm not sure I'd really frame it but essentially the threat model regarding countries utilizing this technology in a really malicious way against neurons citizens for control purpose.
01:13:18
Speaking just wanted to add on that and just to get back to the topic. I think it is revolutionary that there's a huge commercial platform having such funding. I think it's always difficult for researchers to get funding to do research and then we have done a lot of those already that's you know, what you're all interested it's already been published already been credited but I'm not sure how much credit are they giving to the scientists who write to vote and did all the research in last two decades or not and back on this mats point of view.
01:13:52
I Don't. Moment I was hearing about it I just think a bunch of movies close to the shell a lighter the battle angel and bunch of other things where they talk about not just patience right they're just normal people we just want to extend their capability they look they won't ever robotic arm they want to actually, you know store memory better better about way like with that cool ship instead of you is trying to think about actually have a memory chip actually get exact photo of your mind and show it to someone else but when you think about it anyone who have a higher power into the computer system or the control center, it may be able to hack that system that you're Actually using and then actually control your mind or controlling your head, so I always worry about this this wireless or Bluetooth type of this hardware that people can track you people can spy on you people can actually take over decisions for you.
01:14:45
I mean, we already check our pads and then we know exactly where the pad is even though it may be lost and you know, I think we're going a direction where don't we hold a lot of new comedies and new laws and regulations maybe as a band some others were saying that you may be so difficult.
01:15:04
Within US maybe not be the case in some other countries I know some countries in Africa who use Chinese telecom services, they already have the capability to track you down and share your information online, so the you know, political political party can exactly track that person down and you know prosecute you or prison you and they actually have tracking system on all their population, so it's already been happening on domains that we are using right now and I'm sure more we make more things that can be controlled than more things that can be had.
01:15:39
Some retire a little bit go ahead this metal thanks so on the one thing I just took to bring some clarity and on the idea of decoding brain signals. I think it's really important to use words so that people who are laymen can actually understand what is going on where we're talking about breaking computer interfaces and the idea of decoding brain signals.
01:16:09
I think is a little bit disingenuous because, We're still really at the capacity of really kind of like a super polygraph and that you can map a baseline and you can get a control that's related to some idea and then you can have the computer compare do the pattern recognition and apply control to that that is not decoding brain signals of someone in trying to identify what they're thinking is that is far from that in fact that leap is a near miracle.
01:16:48
Now that's a strawman strumming that's all I'm literally looking about what is happening when you map your relativity and you store it and you use the machine recognition system to compare what is happening in real time and you're making a guess right but you're is not decoding brain signals and identifying the content of someone's thoughts that's what your statement is true, but it's a straw man because you're talking about what we've done in the past not what we're going to do with this new technology.
01:17:23
We're just. What I'm talking about.
01:17:33
So the reason why what I'm saying is true is because the the the the research that so I'm currently writing the book about all the research that has been done that it is that that is that we're going to use to apply to decoding thoughts with this technology genetics, yes, you're going to use opportunities to I guess to it and so point and I'm looking forward to it.
01:17:56
I'm very excited for it in the meantime though well from the desert if I could just add real quick. I think when we talk about decoding the signals, there's different levels of signals that you can come up with. Right through decoding and EEG activity sure you're going to get sort of broad scale different regions that and then picking up their activity but if you're talking about sort of like a multiplexer rate type system, you're going to be picking it up the actual individual neurons within the region that you're you're sampling from and so when you start to the code that you know, you you have multiple different probes that are picking up a number of different neurons and sometimes the probes, you know, an individual probe is going to collect activity from you know, three, four, five six neurons that are around it and so you have to be able to.
01:18:43
Take the signals and you have to be able to sort all of those different signals in order to decode like what activity is coming from an individual neuron versus what is coming from a number of different neurons and so there's different techniques that that you can employ in order to literally I guess decode the signals that are coming in to figure out how many neurons that you're recording from and where are they in relation to the other one so some clarity around the term the clutch by welcome to a stop sign and I agree.
01:19:15
If I woke up to a stop sign and I read STOP I decoded the symbols right but I decoded the symbols based on an installed ability to identify those symbols into greed a word right, but if I'm not in English speaker, and I've never been taught English and I walk up to it and I somehow decode that is said stop that is the magical gate that you're trying to cross and that's not been proven nearly not even remotely well.
01:19:48
I think what you might be. Talking about is associating the activity with an action or with a thought. I'm trying to get clarity around the term decode being brain signals and I'm saying that it's a map into store some interpretation into read into pattern recognize it is not decoding in the sense of putting something as somebody's head in reading their thoughts, so they could love to say like manuals point and this is a bit of the person knowledge but like for anyone who's hearing this discussion and is, Was relatively, you know new to these concepts I think manual raises a decent neurons point, which is this that if you're walking away thinking that it neuralink or this technology currently as it's you know, with the monkey video example, it's not reading the monkeys purest thoughts in terms of what it wants to do it's inferring it by training a neural net that every time these particular brain patterns are occurring in this part of the region the monkey seems to want to move the joystick in a certain.
01:21:00
Direction and it's predicting or inferring intent I guess and that's not the same as the monkey is actually thinking of I want to move this thing I guess and and so is it? Yeah, that's what the useful nuance point for anyone listening it's a it's a really important point because the ultimate goal like I think eight suggested of getting to the point of creating able to actually do augmentation or as we call it intelligence amplification does not require one to be in the brain that's the point of those try to make sure yeah, all right go ahead later.
01:21:41
So what do you think is the real decoding?
01:21:48
What is the real world? Decoding would mean that you walk into a room and somebody has some sort of like EG sensor place on your head and whatever you're thinking starts essentially just going into some you know interpreter that says what you're thinking and that's impossible it's important because the reason why it has to be made clear that that's impossible is because behavior environment data is not just about like my privacy in the sense that oh yeah, it was my thoughts or it was my email it was like behavior metrics data is your nervous system.
01:22:26
And activity in real time now if if some company or some whatever government figures out how it's a straight decode behavior by magic everything that overall data all signals that come from nervousness of our behavior environment, whether you want to call them, you know, EEG or you want to call it heart rate you want to call it just freaking fish recognition whatever it's behavior of our measure but to be able to suddenly.
01:22:56
Without the reference the decode behavior of our metrics, all right and then use it is the ultimate. Control of humanity it's the ultimate, it does nothing beyond that. If I cannot think of a question regarding like you're you're you're focused on decoding I think my statement when I was asking about whether or not a country or or entity could point to this technology and malicious way of it and individual or group of individuals.
01:23:25
I think it's not whether or not it can be decoded in real time and and and controlled or sort of bi-directional single amplification that you mentioned but more so to the point of analyzing and I it just sort of like the the precedence on this is that black mirror episode.
01:23:41
I think most. Of. Probably seen if I could take a package of presupposition but it's where the the patterns could be analyzed because I think Meek had a really strongly about the future uses of this and there is one hypothetical where if we're talking one analyzing and like recording short-term memory and all of these neurons and activity going on in the brain and if we were to record that like we saw in that black mirror episode that sort of where my head was out when I was asking the question about hypothetically, let's say without you know, proper constraints or ethical guidelines, like we would see maybe in the West, Would a is there sort of like this idea that or threat that we hear the president the good music is just quickly if I can respond to him the good news is yeah, we're not able to decode green signals, all right and if we were it would be extremely dangerous but here's the bad news we can do this not that true imagining without the according brain signals, we could do it with basic behavior of our metrics whether you typing on the computer whether you're essentially taking whether I'm taking two or three shots of your face and in ten.
01:24:51
Nervous from your camera phone, whether you're active further through the accelerometer and the lower God help me you give me your heart rate. I mean, I used to work for the state department doing sci-ops Afghanistan 12 years ago and I figured out back then that I could do I could not be a neural dynamics between your based on your heart rate and that's before you even get into the content of your psyche so I mean, it was possible today, it's sublime.
01:25:21
I think so just that that analysis that you just to determine right like the telemetry and all these other biometric indicators that's interesting to a point but you're still running on a assumption basis, which is you're assuming that these various data points are correlating to a hypothesis that you are what that's the thing you just like I just said again if you're looking at it in terms of me trying to identify what the content of your psyche is, you're missing the point I can make better inference based on your neural cognitive dynamics to decide what decisions you're going to make out unnecessarily care.
01:25:56
Content of your psyche that I did you have a comment and an in between us because I haven't heard addition to that after yeah. I basically same thing as um, well menu you you kind of just like answering yourself is like essentially you're asking a question of like it's correlation equals to conversation right but then how much is it matter right like if I could if let's say if you're injured it only matters if you're in your scientist like oh, I want to know if it's signal it's the cause of this behavior, but it doesn't really, Matter if you're trying to make a commercial product or if you're engineer like as long as you have a working model that like that's like can have a high accuracy in predicting something like.
01:26:43
Yeah like I don't see I don't like yeah, I don't yeah yeah very high accuracy right now as what I'm saying, they're at a very high accuracy right now and what we can do first and foremost is actually create bring some clarity to the difference between behavior barometric data in real time because that has to do with literally knowing your cognitive dynamics in real time and if we put if we just throw that data in with your pictures on faceobok, the advantage is the difference between a mosquito bite and Frenchman on your, Privacy and an elephant trampling on your cognitive and your cognition in real time.
01:27:23
Like I think that manuals are getting hit on a really important point here and it's just touching on something that I was like. Kind of explained earlier which is like so we keep talking about the ethics of this stuff and then like the cognitive and the decoding and it's so important to address like we.
01:27:43
What what is the big difference of actually having an intracortical or an implantable electrode system or decoding actual neural signals and all that, you know, because we have so the neural signals are mapped into physical reality through the body and language and things like that so you we already have the ability to map facial expressions, you know, your habits what you write like Google is like analyzing everything you write in all your in all your emails all your searches.
01:28:14
faceobok is doing the same thing and then they're analyzing for. You advertising type of profile and then they're selling it to advertisers like what so I think it's so important to get into when you start getting access to neural data what does that changing in the picture because you can't even talk about that if you don't understand you it's like man you bring in what the current picture actually is which I think most people are probably don't even understand the level and I don't know if it's like sensitive air territory here or something but like you people you know, when someone's watching you they they can like yeah, You can learn a lot about someone just by what pushes right looking at their habits and like more than you can fry them, okay?
01:29:00
More than most people can fathom yes you can learn so much more yeah that's what I'm saying is like what does neural so that's what I'm just saying like we have the neural your brain is already mapped to the outside world, you know, that's your body and it's your voice and and what you see is it input and what you hear and it's like this is all available right because you don't need a brain computer in your case to do that the interface is the sense organs in the motor motor control, so that's I think that's the such a key point and it's, It's kind of related also kind of this whole ethical thing is like everyone talks about what's what are the ethics of of like what we can access and all that but like why don't we also talk about the ethics of that's actually currently happening which is breaking ethical and privacy guidelines yeah, so those are just some interesting can I say, can I say something about?
01:29:57
Yeah because I just if for those people listening like I what I think just if I could try to sum it in layman's terms you guys are saying that the vast majority of the things that potentially could sort of be the equivalent of those what's that I'm at what was that we're getting notes enabled those black mirror episodes where things get really really out of hand those things could occur long before we can actually read thoughts and in fact many of those things in terms of.
01:30:32
Applications happening through smartphones and and will continue to inch away and so we could begin those debates and separate that from the very specific advancement of neural link is that is that a fair summary of that. I think yeah, I think that's a good that's a good point what I would what I want to say is is a couple things one is that that so so there's that you can look this up on Wikipedia, it's called effective computing and it's basically the idea of using multiple sensor modalities or You know just basically analyzing user behavior and the behavior on a computer and and and making predictions about what it may be about what making predictions about the user that could be how the user feels or what their intentions are this is this is a kind of decoding and and really what I want to talk about is decoding for a second, so you know, when it comes to the to the the Utah array or to the the the neural link device that we've seen I'm sorry.
01:31:35
I meant brain gate or the monkey in the neural link video where the word is basically. That is like recreating the brain gate where we're decoding the monkey use intention based upon reading neural patterns to the intention of where the monkey wants to move the cursor and so that is a certain I would say let's call it a certain resolution of decoding that you can get with brain gate now when you increase the resolution and the the the sampling rate and and and basically the the the amount of data that you can collect from the brain so that basically what you can recode or reconstruct from that data is a vastly high resolution.
01:32:17
The basically it's like it's like going from not only knowing what the monkeys intention is but knowing of where where I want to move the cursor but but knowing very precisely where it wants to move the cursor and maybe and maybe knowing a little bit more about what the monkey intent so so I you know now when you're when when the monkey is maybe the the computer is doing eye tracking on the monkey and there's a bunch of things and the environment and the monkey is seeing a banana, and and you're and there's some neural activity that that.
01:32:52
Neuralink is picking up when the monkey thinks about a banana and and so the monkeys still doing the curse or thing that's it, you're it's an early because it's predicting where the cursor needs to go very accurately faster than the monkey can move the cursor when it was moved when I was using the joystick and and the the neural link is noticing when the monkey wants a banana or an apple or a grape this is this is high resolution decoding and when we and as we the the the thing is with neuralink is such hugely forward in the resolution and capability that what higher resolution decoding is like, This is when we can we can begin to like, you know, we know that the monkey is thinking about a banana and we know the the the curses over there and and maybe and you know, you put this in human being and maybe we know what word a person is is thinking about or or what which picture a person is thinking about and so the resolution of decoding matters a lot and and that's why it's this is different.
01:33:51
A very good point, oh go ahead man, sorry thank you the resolution between I guess like a regular EEG device and neural link is astronomical nobody is debating that it's insane but it's still in the motor cortex and the mortar protects obviously is extremely favorable to mapping so that works perfectly for people who have I don't know people who.
01:34:26
Are disabled and would like a way to interface with their computer if they don't have you know, the voice or or or tactile what or something like that so nobody is questioning the value there. I'm I'm just trying to bring a clarity around what it means to read your mind and what we're saying is if what you want to arrive at is intelligence amplification, it turns out you don't need to be in the brain at all because there's more than enough information in real time, especially if you have a wearable device on or like, Your face and it intermittent times, you know to give you the opportunity to cultivate that closed loop potential and improve your neural cognitive dynamics and with that your decision making and exploration capabilities and learning curve that is possible today, so not not disagreeing with you not disagreeing with you there in that terms of like affective computing and which is the way to describe what you just said is a way to the that what we can do with EEG and eye tracking the limits of those technologies alone just those.
01:35:36
Two hasn't come close to what what is their potential possibility but just those two technology so we we haven't there's far more but but but to the topic of the room I would never link offers is is for you know, the two main points I think about is this is not a technology we can stop from happening and to next level decoding of thoughts is is very very likely and and I if I apologize to people who don't want to hear that but, No no and I think just I want to definitely just a few folks who haven't chimed in yet but they wanted to get their opinions as well, but I have a larger question for you guys here there's a there's a lot of technical experience in the room.
01:36:25
I'm going to use an analogy to help sort of frame the question so when we look at Tesla like we know where the breakthroughs are it's essentially you know, obviously the battery manufacturing capabilities the the ability to you know, autonomous driving is next in the horizon like you can start listening out like the core breakthroughs when it.
01:36:45
Come. To neuro link like what what are the like they're touching on it but I'd love to hear like what like thematically what are these like poor pillars that they're advancing that they that maybe the the built on the prior work, but as these guys have taken into like a necessary step to commercialize is it the the wires and the position around surgery surgical implement plantation the not the resolution if you guys can just comment on some of those core architectural pillars that you know, that make your link.
01:37:20
A particular revolution. I'd love to hear people's thoughts and. You know, yeah, so go ahead. I would say the scale you know the scalability of the product because I think most of the technology like has been mentioned by aim by make by mica and most reputable neuroscience departments will say an academia have most of the tool sets that are available that neuralink is using it's nothing, you know monumental by any means in terms of what they're actually recording and doing in the brain but to be able to scale that is don't small feet and that's what's is very very impressive but not only that but kind of in touch.
01:38:04
Done it earlier too, which I never really thought about but it's it's really an interesting idea is that the ability to handle that much data in some sort of smaller package that's that can be utilized across, you know, thousands or millions of users because even as it is right now, even if you stick one probe into the brain to record let's say 300 or 500 neurons individually it is it's not a fast process to analyze all that data and you need a certain amount of computing power that most.
01:38:39
Residential folk do not have at their hands so it'll be interesting to see how they'll be able to sort of scale down the computing power that's necessary in order to handle that much information that's coming in because you guys are talking about decoding I think there's a number of different levels of decoding and so you know, when we talk about it first, you need to you need to take the signal that's coming from the brain itself needed to figure out what's contributing to that signal which neurons are firing and how are they firing and and then on top of that now you need to decode the activity that is you know, responsible for creating.
01:39:14
That signal which is more you know putting it to an action or a personality and then you need to map it into the person themselves for independent individual differences, so I think the ability for scale and which is why it is very cool with neural length being, you know very much with a lot of musking in the Tesla group.
01:39:33
I think the capabilities of them to scale are definitely in good hands, but that would be what I would think is the most revolutionary thing about it.
01:39:45
So could you do so maybe if you are if someone can elaborate on the scale piece it might hearing this correctly so every aspect of what they're doing even the choice of materials like has been used in in labs that have been essentially working in this specific domain or are they pulling things from I don't know the semiconductor or other industries that historically are more, you know needed for the scale aspect of it but are they bringing another?
01:40:16
Means that traditionally we're not brought into the mix to achieve scale or other other technologies or other applications. I'll give an example like for battery manufacturing, you know, Tesla had looked to bottle manufacturing and adopted some of these practices to to help, you know, increase the production capacity and reduce its costs and achieve economies of scale, so I'm just curious like like what are some of these things that these guys are doing to achieve scale is it is that stuff?
01:40:47
You know any technical advancements there that they're pulling from other places. I had to actually do questions, one of which I've invested at this point, so they sort of related, so I'm gonna put it again I should probably mention. I'm not a bible person. I deal with physics so very far from this but there are two things that should be ignored. The first is I think I really like my first point of you know, how much resolution I'll be getting but there is a very different ethical dilemma based upon how pungible a learned model is.
01:41:22
Replaceable the human is that is if I train a model on say mica and how he conducts more attractivities how does that model transfer to another person right and based on that you have different entries of privacy concerns and ethical concerns now that's a very interesting notion that every very curious just to understand how similarly different brains interact that I feel that is scientific gold in an opposite cell and it's also ethically really concerning because it means you could say trainer model then force your device upon a prisoner of A water or something and use them to interrogate out the entirety of the enemy's military formations using an implantable device that's freaking scary now.
01:42:05
I was thinking that I was thinking about learning and I was rising let's say you have a model which you've learned right now conventionally what you guys probably do and you'll know this better than I do is you wait systems so that your output matrix can be reduced and you have a sensible result but okay, yes itaves the model say I turn on a light to on price and off one.
01:42:32
Or it's one one zero one one zero or one one zero or anything like this you have like a finite state machine and it's very simple in electrical engineering to come up with a circuit which will look at the system which goes to a finite state sequence and directly quote that you've given result now you could either go through a lot of transistors to make do with this in a very abstract system say an x86 system or you can just come up with an FPGA that's a field programmable gate array or an asic and use custom.
01:43:07
Digital logical to strain to recognize a certain sequence and use that for processing so what I'm saying is say you had a reprogrammable ASIC or FPGA in every person all of a sudden the scale goes down to such a degree and there's so much potentization because of the way it operates that I do think that becomes entirely tenable to handle this on a very very loud scale.
01:43:34
I think that's I think that's a really interesting point Regarding scalability though like because we're really hypothesizing about future technologies and if we apply more as law at all of this as well, right there is the application that or the inference that we can make that scalability for processing and increase processing power will go down in size or we'll find some your new revolutionary method for handling all of this I do think though on this note about talking about scalability and all this, you know, large-scale data processing for everyone in The room a general question I want to throw out there is what are the implications of being able to to analyze the big data coming from everyone's brain?
01:44:21
What kind of like what would that mean for like we like manuals made some really good points about the bioindicators how much inference you can get out of those just by itself and those are generally available right now. What happens when we actually can scale and find those communicators right from the human brain what does that mean when extrapolated and analyzed across like, Population sizes.
01:44:51
And the point is you definitely don't need anybody right up by the brain and having them population of cyborgs to get some professionality because you can do it with the existing technology the question is are you going to create laws that protect individual behavior biometrics to make it even feasible in the first place but it would be an extreme advantage yes to arrive at super rationality.
01:45:17
I have a migraine that laws are not a clear indicator of prevention, they're they're not they're they're. More than anything, laws are about creating design spaces. Right now everyone in technology who works with mission prediction is designing in the approach of artificial intelligence, which means that they're trying to create a standalone complex that will be able to humanity and be a benevolent shepherd to it.
01:45:48
That's one approach. Another approach is intelligence amplification. You're using the same resource that is called machine prediction, but you're doing it in a different approach. The individual can optimize our common dynamics in their learning potential, but you cannot do that until you can essentially create a line that separates the data necessary to do the one versus the other.
01:46:14
If you have no line, then everybody's approaching it from a standpoint of artificial intelligence and everybody's using all the data and however, they see fit and you have a land of confusion. Now to catch the point that you are in and it goes right along the line with your question.
01:46:36
What is your working gain from this kind of thing is is the quest the problem of the empathy gap if you create a universal model of human the human mind, whether that model is, you know, it might be universal in terms of how you gathered data in these initially modeled it but you still will have an empathy gap in that it will not relate in any way to the mind of an individual walking down the street like and if you assume and say well based on the science, I might have, Been collecting 10 billion parameters in created this model that it's that is right then you're now in the land of tyranny.
01:47:22
That's a really interesting point that's what that's actually kind of wondering when it and I think you summarized it. I'm not sure about the the the integrated empty gap is what it's called, but that's interesting because I was wondering regarding how the individual neurons firing in one person's brain if they would be almost identical in another individuals during based on and you know, everyone's individual neuroplasticity and everything but that's what really interesting point.
01:47:49
I just wanted to make a point which is sort of on a different aspect of the neural links impact on us and vice versa. How is the human mind going to cope with something being implanted into it which is neither the self nor the imagination what keeps us from going down this like weird schizophrenic out of touch with reality thing once something's exists within our own minds which don't retain to the self don't pertain to ourselves what will be the impact on our own psyche is my question and if there's any, Neuroscientists or psychologists in here who can tell me if the brain will cope with hearing voices all of us from the scenario thing.
01:48:35
Don't we already have something like that with like had any differentiate between neural interface that allows you to like control the computer and like a keyboard and a computer that like, you know, having access to social media or all kinds of stuff like that. I think the difference for me is that the sensory input and sort of action output it's clear here, these are tools like our eyes, mouths , hands, these are tools that we've evolved with.
01:49:03
For thousands millions God knows how long so when something like this gets implanted to the mind and body out of nowhere how how are we going to be able to handle that so I think is raising a point as in we already have the center of modalities five modalities to that say but having this implant will be our sixth both modality that is not part that has not been part of evolution or development as suddenly implanted however, how is our brain or mind or our consciousness going to cope with interacting with such external.
01:49:33
I guess I guess external hardware that's in our Brain. And just thinking about that I think for people who require such equipment that's saying the medical community people are more inclined to use it but for people who just want to enhance their capabilities. I think that's where people are gonna be murky about whether they choose to do it or not, but for the medical community.
01:49:58
I think it'll be a huge push in advancement. Also Nikki can you elaborate cuz like well that's all I just think I worked out a project called brain gate that was a motor interface that we actually did an implantable motor like enabling people to control computers robotic arms with thought that's pretty much just like they could with their arms so they could think about moving their hand their arm and then the computer will actually do what they want.
01:50:26
I was wondering if you can. I think what you're trying to say is like in the case where you have an auditory press thesis that you're getting sound into your brain without the traditional sensory organ like the ear. I think that's weird what you're saying. I'm just not 100% clear like you know, what you're what you're getting at with regards to a neural interface because I can really interface does not necessarily imply that things are you know, working completely differently from the current system, for example, if you don't have an arm and you get a neural interface that decodes the hand.
01:51:04
Ler movement and then you get a prosthetic arm connected to your body and you're controlling the prosthetic arm. I mean, that's not that's not different from having an arm in most regards it's just a replacement arm or if you have an auditory processes or even like let's say a cochlear implant or a more neural interface auditory processes, so you have a microphone and then you put the sound into your brain that's not any different so I I was curious, can you elaborate on what you're getting at?
01:51:37
Um sungu actually did a pretty good reiteration of my point here . It's the whole concept of having another sense added to us in the case of having a robotic arm be controlled directly by the brain. I can see why that was a bit more sort of transference with the skills our brain and body already has for things like moving arms. My anxieties are more directed towards what it would be like to have a Google bar implanted directly into our mind.
01:52:04
I can actually answer that with a real example, so there's this there there's neuroscience. Ific his name is David Eagleman, he was Texas based in Sanford he's San Francisco Bay area base now but he um, he wrote he's written a bunch of books. I recommend that you check out his books, um, and one of the things he talks about are basically alternative kinds of of neural interfaces that implant information into your brain, for example, there's been there have been tongue strips that have electrodes on the tongue and and they're connected to a camera and you give this to someone who's blind.
01:52:44
And they can begin to make out an image from the data from the electrodes on their tongue and that are coming from the camera so it's took so it's like imagine that every pixel in in the camera which represents the images being captured is causing a corresponding electrical stimulation on on a flat grid that stimulates their tongue and the person is able to see basically accrued picture of reality via via that stimulation so the image from the camera is going into their brain because there's that because they're tongue is being stimulated this this same thing has also been done.
01:53:19
When the back strip of electrodes a really large back strip where people were able to see through their back so the camera was plugged into you know, they the the the pixels and on the camera corresponded to a simulation on their back and their brain turned that into an image so that's an example of how you can plug in data into the brain and so David Egan for further goes on to talk about how you could do that with stock data, you could you can send stock data directly into someone's brain with the with the similar idea now if we if we bypass the tongue in the back and and then we just you know, basically directly.
01:53:54
Stimulate neurons and it the idea is like okay so if we can by watching a person eat broccoli enough times and watching many people eat broccoli enough times watching their neurons and watching in just the right places so we noticed just the right there's right temple and spatial patterns in neural activity that that corresponds to broccoli most of the time then the idea here is that if we could if we could then recreate those patterns through brain simulation if we could recreate the temple of spatial patterns by stimulating neurons that we saw the temple.
01:54:29
Spatial patterns that we saw in medical imaging that corresponding with broccoli if we could recreate them inside someone's neurons then that person might perceive broccoli where there is none there's there are experiments which you can give someone special brain simulation and they I think it's transcranial magnetic stimulation and they might see like a light there's examples of you know, I guess I guess I mean, you know, there are there there are many examples of altered brain activity where someone takes some sort of medication and they see things that are not really there, they see patterns are not really there who hallucinate hallucinogenic medication for Example and there's also examples of you know people who hear things are not really there and in you know, I guess schizophrenia has a topic that that we don't really want to go into but but but you can sort of like consider what how do those people deal with it and it's it that particular in that particular instance it's schizophrenia is very devastating and like so we don't really want to go to to that topic because it's kind of sad but um, but yeah, so so these are check out David Eagleman's work and in that may really be in the area of your.
01:55:39
Interest.
01:55:43
It's going to also ask if you know, if David Eagleman conducted these experiments with conducting sorry putting these. How to call it putting the input directly to the mind through the tongue or the back or whatever of images and the mind rendering these images and actually seeing it what it worked with people who are born blind or are these people who are just they had their eyes covered they want blind with to do with that but at the same time.
01:56:10
The schizophrenia aspect I'm hesitant because you can't always tell ahead of time and I'm nervous if they start putting these implants into people who are predisposed or anything like that could be an entire disaster but that's it two cynical. I think the potential for mistakes is big.
01:56:27
I just forgot what the previous question was.
01:56:35
Yeah the people who were being tested on did they always have vision did they lose their vision were they born blind okay so I can answer this question and a general way you but you want to you probably want to look up David's studies for specific answer but the gentle the general question is that so what happened is there have been experiments on rats where they've taken an eyeball out of a rat's eye socket and they've plugged it into the part of the brain that is the the called the consider to be the auditory cortex and the auditory cortex of the brain.
01:57:10
Which is supposed to be for processing your ears learned to see so the idea here is that the the neocortex is a general learning algorithm you can plug in any sensor into any place and yeah because because in this the reason is because basically it's as if the neocortex is is has the same sort of way of functioning no matter what it's functioning it's like like you plug in any sense and it can process any sensation, it's like a general alert learning algorithm.
01:57:42
To add to Nikki's point about people who are born blind because I did some work on people who are not born blind but became blind after even a lesion or surgery or accidents in their primary visual cortex, but I think you'll be interested to know the work of I'm Eddie from Israel I think is a wiseman institute or maybe tell of each but he works with people who are condemned they're born blind, but how do they actually learn to navigate through space and see is what he does is that he actually trains these patients.
01:58:17
Were blind people to use cameras that actually read out your area and then send very specific sound timber pitch tone whatnot, so you would associate different type of patterns of sound through space color depth and objects and more recently he's been shown that these patients can actually differentiate between color of red green and blue or light if this is bright this is coca things like that and they could literally have these blind patients were successful in learning easy over two to three years can actually rub a walk through maze and if You actually run FMRI and other type of embracing data I bring the imaging techniques on these patients there, they've never had any visual input through their visual system so technically their primary visual cortex will be just not working assumed but when they actually learn to navigate the visual world using sound as an association, he actually saw a representation of space in primary visual cortex, but their input was auditory system right so.
01:59:24
Our understanding that visual is specific brain areas are very localized and specific may not be so true that certain areas we call visual area are not just visual area there are other audio visual multi-sensory integrative neurons and populations that actually interact with one another so I highly recommend you read into a magisborge if you're a very interested and some people like project precaution by power sinha from MIT, he worked with patients or actually population in India who easily after character can.
01:59:59
Gain vision but they are it's so you know in such remote part of the world they don't even know about the surgery so when they're born, they literally have no vision so they take he takes all this doctors to have cheese performed a catapurvy and suddenly after nine years you gained this rich information of feature about the world and these people have not having access to vision in their critical period are now suddenly having all these information coming through their eye, so of course, there's some transition that happens individual development that they have problem with occlusions and you know detecting certain.
02:00:34
Lines of what not but with enough treatment these people these patients or these children can actually gain as normal site as a person who was actually not born with such eye problems so I think there are enough interesting research going on right now on people who are born blind or could be you know, normal but did not have a surgery.
02:00:55
I also wanted to like, I think I'm starting understanding the concept of like if you just start getting access to tons of data and information and you know, I think a lot about your showbox at the description of consciousness and like, you know, like what if you can just your BCI is like accessing and helping you control your own attention or like direct access to memories and like filing through the memories and filing direct information or cognitive stuff.
02:01:25
I think that's really interesting question and I, I it's kind of like what happens like is ideal and then what happens in that situation like does your brain just like spark to life with like this new dimensional types of abilities and I there's another area that's related to this that I'm interested in I it's hard to describe it but it's basically looking into almost consciousness but like psychic phenomena and things like that that would where you're talking about schizophrenia or psychosis, that definitely would be a negative issue like but there are other there are other.
02:02:04
Things that are in like alter-state consciousness like we already people do like LSD or something like that and access different areas, what is the BCI or neuro technology can actually either go in a complementary way to using psychotropic or doing educating similar capabilities with a technology so I think that's a future area of neural neural interfaces that is also I think interesting or possibly related so I'm actually found that very interesting because I have a certain application.
02:02:39
So let's set up a state space right so say you have solutions. I'm gonna use match terminology, please some of the CSP below here break it out into the language you guys actually use so say you have a real number right anyone any real has a cardinality the space of real life has gotten out of the continuum now so you have r2 or r3 or r exponent n where n is a natural number all of these will still have the cardinality of the reals, that is the new.
02:03:11
And if I wanted to say map all of say r2 or to r, you know r1 I can essentially using a modular system map all of r2 on r1 because i can split it into parts based on congruences there is essentially a linear tradeoff between harmony different term states i have and how accurate you know, i have to be if i have a single finite double as a variable type.
02:03:41
So my question is is there also a linear drop-off between how many nerves I am looking at and when they're firing, like if say a nervous fire is on or off at a certain time. Is there a linear trade of an accuracy of the number of nerves you have for the same state space and secondly if as mica says these are all spatial temporal signals.
02:04:03
Does that not mean that when you send them in every input has to have a certain nature that is say I have ratio temporal reaction alpha which corresponds to me tasting of banana does that mean for you to induce the taste of a banana in a subject you need to be stimulating for a time period alpha so essentially we are limited by how much time each subtask takes to process or is there say harmonic system to these spatial temples setups where we can just be these.
02:04:38
Up by an integer multiple or something and of course the objects work around I see over here is we don't send in a read pulse which is this is what a banana tastes like is like we can try to send them a right pulse which is this is my bodies reaction after a piece of the crack and I'm trying to understand how these things play in terms of complexity and getting around this time limitation.
02:05:11
I think if there's again neuroscientists who can explain some of the neuroscience basics like the systems neuroscience. I feel like you're asking. Math type questions, but I'm not sure how they go directly into, you know, systems neuroscience.
02:05:35
Or yeah, I don't know someone who has a neuroscience background. I. So, I mean, this is not it's specifically cognitive neuroscience but it's also so we're just so the topic we were talking about basically the site the like the precise mechanics of decoding temple spatial frequencies.
02:06:00
I sort of attuned out a little bit with what cash off was saying. It's just that some of the things that that you're talking about are beyond a little bit beyond where the research is at currently. That's my. That's my opinion from what I've read.
02:06:23
Okay. I'm sorry again. I'm completely outside of I've never taken organic so everything. We're here is creative right so cache is it is that odd did I say your name right oh excuse of put your here okay so um when it comes to neural encoding there aren't any at least like to my understanding it's basically like a classification and clustering algorithm, but if you're trying to recite model the human dynamics response, like for example, the the brain actually regulates the blood flow and like into.
02:07:04
The blood and like it regulates the the blood flow visa fee controlling the cerebral pressure and there are state space equations that actually dictate that you can actually create a transfer function that can essentially model the the input as well as the response of of, you know, grain actually regulating the server respond and like serial role pressure and but these are like highly granular responses like that, they don't actually really apply to.
02:07:39
To say like decoding the brain because you have a degree system mostly if you're trying to decide wheat out the neuron's response because you're you're you're just counting the spikes and and neurons have certain discreteness and the way they they fire so I don't know if that answers your question but oh and how they're not and you know clarify them so is there like a time series like say say I'm modulating kind of certain rep rate.
02:08:10
I can only send in so many bits of information in a given second. So is there like a limitation to how many fish you can send in for you know, given spatial temporal pattern. Right so this is highly constrained by it and the firing pattern of neurons so if I'm not mistaken the considering the despite the people organization hyperb and hyperborealization of a neuron you're looking at two to five milliseconds so that by itself would limit the amount of information that you can gather, of course, the nearly group has actually has more sampling rates and that is to create certain being in their ability to decode but on top of but like, If.
02:08:57
Biological point of view like you're looking at a very limited time which is like two to five milliseconds and that's widely constrained to your ability to get like a continuous time to respond because you won't be able to get that that answers your question yeah just sorry I'm thinking about our time just want to clarify so what you're telling me is if I have a perfect good drive signal and it's nice and triangular perhaps you're telling me that with 200 to 500 kilohertz.
02:09:26
I'm gonna match the behavior of a neuron. In fact, there are mathematical models. I don't know if you heard about the model which was the earliest adoption of at least like the earliest model which is basically a partial differential equation which dictates how neurons fire and they were able to do that on a giant squid axon so what they did was that like the the implanted electrodes into the axon and try to see the response stuff the different or at least I can mean a measure the the the depolarization and hypochlorization and, That to the a mathematical equation which while it is successful and has been successful ever since I don't think there's there's there's been any modification in using that kind of model that to put it be like in a simpler terms yes, there are mathematical equations that you can put to model neural response and it's also going to be constrained within like.
02:10:36
This I think with the time series I don't know if Nick wants to jump in front of me because then he's a very good neuroscientist but you know with in terms of a time series. I think what you're what you're thinking about with the neural link is basically watching one neuron fire after another and then another one is so from that like the time series is is constructed probably constructed from that in terms of yet two to five milliseconds, but I think the whole neuron, you know, after it fires it has it goes it's inhibited for a while, it doesn't have any.
02:11:11
Activity in the full reset time. I guess. I've heard it was closer to ten mill seconds before it's sort of ready to go again.
02:11:19
Yeah and I guess that's kind of my pay grade as far as the data perspective of that. I don't think the time series is ever going to be necessarily limited because even with the two to five hundred hertz sampling you're going to have more than enough to capture the resolution of the single neuron and its different biophysical properties.
02:11:43
I guess if you will because yeah, it's you know, the average action potential for. In spite of this neuron is going to be, you know, still like one millisecond which is usually going to be able to be captured but after that you're going to have, you know, two to five. Fraction where it's not going to fire again.
02:12:04
That. Start nice way so. I'm thinking I think you know, technically we sample for.
02:12:14
500 to a thousand hearts and that's very much within the capability of picking up, you know, most if not the entire action potential itself. I try guys you guys are missing and this is this makes my night no I love this stuff can I ask it a question kind of can I ask why you were inquiring about the time signature just because with your background you wanted computing and how much processing it was the time thing that the time issue related to sort of an inference that you had.
02:12:48
Actually that does one thing. I'm wondering if this whole system is created by how much time it takes to browse the information in which case there is an upper bound to how useful seniorly ever could be so you can't sit downward and I move. To cause you can't say have a single scene which takes you half a second to digest into 24 into 60 into 72 for a 72 minute movie secondly and this is fun so I just hung up on this quit it's a superconducting fortunate nation of fundamental parent spies and these consents my mute magnetic fields, like I you can tell if a person's ideals are open up but so we're going to reach a point where we can essentially judge people's what the muscles are doing what the neurons are individually doing in different parts.
02:13:36
If people are wearing say a suit which has superconducting devices on it and that means that you can get tons of information out and I'm wondering if it's possible to create a feedback system which would be at far with it and for reference superconducting systems have very low impedance so it's not uncommon for them to work at like an RF frequency range so it's super super fast like 40 gigahertz is perfectly normal so I was wondering if you could have 40 gigahertz I/O to a human brain and that would be kind of yeah, let me pretty awesome.
02:14:09
Well yeah and I don't I don't know if it helps at all or if it's completely off point but especially in terms of if we're saying like augmentation of brain activity like if you try to speed up the natural processes that are already there the neurons tend to very quickly go into it's called paroxysmal block where they don't fire anymore, so if you try to speed the up the rate of those firing of the neurons in order to stay, you know, take a normal process that takes one millisecond and condense it down into a tenth of a millisecond in order to fit more spikes within a given unit time.
02:14:44
You're just going to end up sending the neurons into a block where they're no longer going to be firing anymore and so it's sort of you have to at some point maintain that time series in order for them to repolarize to to fire for the next action potential if that makes sense, okay?
02:15:03
Otherwise you could result in seizures as well you start firing driving these groups of neurons at certain frequencies, then you're going to result in seizures. I do a lot of work with transcranial magnetic stimulation and know if we start getting into higher frequencies like E, you know even like 20 Hertz but we run the risk of causing seizures and even normal individuals and to kind of get back to something that Nikki had talked about earlier with TMS, we do, you know brain simulation.
02:15:38
Like over motor cortex and someone I think Michael mentioned it earlier or Donna Roberts my friends online, but in all of those instances that I've you know done it to myself had other people do it to me many times. I mean not always new that I was not doing that myself, that there was external stimulation to my brain causing those movements or those flashes of light or speech arrest or whatever it was that was happening, so I think that there's still even even with a higher resolution of a neural.
02:16:13
Link that we're still going to be able to sense that this is not me that it's not my brain that's that's doing this and I think it kind of gets back also to something that was talked about earlier is actually decoding signals and are we really doing that and ultimately in order to do that to really decode someone's signal so it's going to be that that's equivalent to being able to decode consciousness, so even though we might be able to put in to someone's brain stimulate the taste area of this tasting light broccoli or the visual.
02:16:48
Area this is a green object that I'm eating it's it's still this the sensor like I think the famous question is what's it like to be a bat, you know, you you have this physical properties you can hear at higher frequencies whatever but you still but you still don't really know what it's like to be about and in it and so I think it's the same thing but we're not sophisticated obviously far off from being sophisticated enough to be able to decode signals to the sense that we would you know, that it would be consciousness.
02:17:22
Are that we would mistake it for ourselves. Very very well said and I'd like to respond to one to every different take so I agree with everything that this and Donna said. It's absolutely true that there's a limit to how far you can push neurons and I learned a little bit today from what Nick said.
02:17:45
I didn't know exactly what with with the process was there and really great to hear from Donald's experience my thought is is on the computational is sort of like in the computational side of things like so so like if there's there's a fundamental limit to to how much you can stimulate one neuron at a time.
02:18:02
But but when you're talking about the brain and we have 86 billion neurons if if you so imagine like imagine you're trying to transfer information to the brain and if you just doing it through one neuron at a time then yeah, you're gonna have some hard limitations on how much information you can transfer but if you're able if you have a high bandwidth situation and you can talk to a billion neurons at a time then then you know, you kick I mean, just imagine like, you know, just think about that, you know, how much how much information do you need to send?
02:18:33
And for each frame to get a frame of the movie at a rate that makes sense for that person to see that movie right and so if you can simulate many many neurons simultaneously, then that basically like each additional neuron or it. I mean at some point this like the amount of information you're transferring is multiplying per per per per each new unit of computation it's so you're speeding things up by doing things on but by map by sort of what's the term?
02:19:08
Massive parallel computation.
02:19:15
So to add on like what keshav actually asked earlier. I found these mathematical approximations. I had to discuss some of my earliest electro materials but you can actually apply the drag delta function to discrete dirac delta function to essentially model how neurons spike and that is that gives you like some kind of theoretical approximation is to how neurons would actually spike in it in a more distributed matter to I I hope this answers request or leave.
02:19:50
Even like gives you an insight on how to use mathematical approximation just just for others in the room could you kind of help us I guess map out so based on what you just you guys discussed what what are there any upper limits to the way things are currently being designed and with that mean that there's certain applications that in order for us to unlock other technological breakthroughs need to occur could you guys maybe just translate that?
02:20:25
Back in layman's terms for everyone in the room. Um, I guess maybe I can go. So um, to catch up, there is there 's there's there's there's there's ultimately it is true there's ultimately gonna be a bandwidth and from limit to how much you can how much information you can physically send to the human brain that's as crazy to think about.
02:20:51
I don't know what that possible limit could be but yeah it there ultimately has to be a bandwidth rate limit, but it's also possible that we will before we even get to that that we could figure out. Something more fundamental, which is some imagine if your brain is a network because there's a book called networking the brain it's a or wait, let me look that up but the book is let's see networking the brain.
02:21:25
And maybe maybe I have taught no networks or the brain so networks of the brain oh by Olaf's born he's a neuroscientist so this book is really really cool but imagine the brain is like the internet imagine that you're you're you're not you're you're neural circuits in there in in your class, you're the clusters of neurons in different neural columns and micro columns are that they connect to each other across the entire brain.
02:21:55
Like like a network and and and so the neuroscientists Olaf sports he tries to use graft theory to to map those connections and to sort of figure out how they work and and and some of the things that some of the ideas of how the brain is network you yield that some of the research yielded something called the the discovery of the default brain network and which is another thing called the rich clubs and in these are like basically like the macroscopic structure.
02:22:30
Is that connect different brain areas I guess you can think of that I think of it that way but um but but what's interesting is like well what is the what is the specific protocol the networking protocol for communication between one one part of the brain another part of the brain and and one way you can you can think about this problem is like it's just considered networking protocols in general with you know, we have our computers use HTTP right and so we have we have a whole bunch of networking protocols that you may have heard of there's there's there's there's one called UDP or use it user data protocol.
02:23:05
Um there's one called TCP which which which TCP is an old one and it requires that you're the two computers do a handshake first before to make sure that the other you are going to make sure it takes a TCP I gotta make sure this other computer is really here before I start sending information whereas with UDP I just throw the information and I don't care if anyone else is there so so so that kind that's what I mean by network protocol does a neuron need to know that another a general cluster a neural circuit doesn't need to know that another neural circuits there doesn't need to do a handshake first.
02:23:40
Before it throws the information or just send the information and and just like the way UDP works and so if we can figure out the particulars of the networking protocol what's being sent what the with the handshakes are if there are enhance shakes if you know how that how signals are received like these are important things but this it's important because if we can if we can build artificial neurons that can talk to that they take this nerd and this decoded networking protocol and allow a neuron to talk to an artificial neuron, well the next step is building artificial.
02:24:15
Cortex so let's say it's a person gets a poll through their brain and they survive but they have missing brain tissue, right so we could put some artificial cortex in there that replaces their old old cortex and now they have a brain isn't suffering any deficits and it might be working better but we go step further and and say, okay, well what if we took of a completely healthy person and we we put a we gave them a another cortex and caught the external cortex right we just put it right outside it's like a hat you wear this hat and it's full of artificial brain tissue and it's network.
02:24:50
To your actual brain but it gives you like another gives you like a second neocortex right you just have another layer of brain but it's artificial and it works the same way as your original brain and what's cool about it is that um now instead of having like let's say that you have a hundred and sixty thousand thoughts a day and let's say that you let's say I just made that up right but but but now you have ten times as many thoughts per second because of your artificial cortex and so that's so for you one one one idea that I had as well, maybe if your brain.
02:25:25
Is processing information 10 times faster What if that means that your experience of time changes What if that means that relatively speaking if you're able to perceive your world 10 times faster that is time for you slows down. And in and and that experience basically means your days are 10 times longer.
02:25:44
Now before you say well, that's a bad thing. I don't want to wait a second to think of all the things you could do in one day if your days were 10 times longer. That is like getting 10 times more life. This is like getting 10 times more time on Earth.
02:25:57
That is the kind of thing that I'm Excited about. Basically like I really clicking Yes I'm very happy about you man So my main concern was yeah initially I was thinking there is an upper bound to how much information you get in court in a given time frame and if there is there's an upper bound to the usefulness of any system that allows for direct input to the brain to circumvent the senses and to connect us perhaps to the internet.
02:26:25
Now that I'm thinking about it, I've started to consider whether there are workarounds and the best work around I think of was that we can preprocess the data. That is instead of sending what is you know, The sense of a banana in your mouth or the taste of banana we can send the information How does the brain react after you've had the planner Our secondly is a case of parallelization because a I think quite a few mentioned this of getting Google mentioned this that is possible to train one cortex to function as another form like the auditory context cortex can process visual information and as you know, when we we can smell something taste something feel something feels something and see something in the same time.
02:27:05
That means we have five different strands. Why can we get information? So say if you were to parallelize how fast you can see an image. And having all five processing them at the same time it will be very interesting to see how much the humans could perhaps say find the difference between two different pictures which have subtle changes.
02:27:23
And to Micah's point of view what is a very interesting cause I looked up the model used to just to model how you can help steam model and it describes things in terms of limit cycles and attractors. I think and that is basically the vocabulary views in nonlinear dynamics and chaos which by the way is how it was learned about this stuff because one of my best professors.
02:27:47
I love this man. Is looking into how the human brain is essentially modeling synchronization and we have found that with a oscillators very often of the electro or optical kind that these graphical networks or networks of interconnected things they have a you can infer the geometry or connectedness of the network based on how the oscillate and vice versa would decent extent and also that the geometrical setup enables these oscillations to create a network.
02:28:23
And I'm kind of curious where that leads in terms of safety and memory being a product of network dynamics over time, yeah. I think she might be making assumptions though about the limits to the interface. I mean, I don't think there's necessarily a limit like technology. There's nanotechnology already, let alone members and micro like I think it's going to be possible to get you know, hundreds of millions of real-time interest input output directed neurons and whatever else is going on in the brain.
02:28:59
I'm just saying in the over talking theoretically here like there's not. We're not necessarily limited on that front. Yeah I don't know I internally like the information yeah go ahead. I was just going to say that you know, a lot of these physics the the folks from physics they they ask they're they're really asking questions that are a little bit disconnected from from reality like completely disconnect from reality no offense catch up but as this is a sort of normal so like in theory like a theory, like there is a potential like ultimate limit it's just whether we reach that in this lifetime, I I don't know it maybe not.
02:29:40
Quite a question so I had to leave for another meeting. I've learned a tremendous amount from you guys that my brain is stretching. I'm a layman. I just have a quick question. I like to think of another room which is basically for all of those of you who are in this industry02:30:00 I'm the original question was neural link is a revolutionary or a tragedy. I'd love to hear your personal breakdown, like what percentage do you think it's gonna be revolutionary what percentage do you think is gonna be trapped a tragedy so just, Like okay, it's gonna be 50 50 is gonna be 40 60, I would just love to hear from you or perspective because you guys are deep deep deep into this tech and for a layman's perspective.
02:30:24
I would love to get that kind of pulse, that was my question Sam and I'm done.
02:30:35
Hear you more specifically. A person using the accessory will be revolutionary. What percentage do you think is gonna be a tragedy?
02:30:54
50/50. Well. I mean, I guess I can get a lot to think about. I think in terms of the technological and kind of human there. I think there were I think we might be approaching a new a new and like era of human development which which is gonna include like direct interfaces with it's not even about direct because I I think so it's important to like differentiate what is happening here when we were talking about like okay if you're okay, first of all there, they're two major.
02:31:37
Aspects to neuralink When is the medical treatment aspect which is just like alright A person who is paralyzed will be able to like move a cursor on a screen or like move their arm. And then there's like augmented error like a person who can't hear can you hear better and things like that?
02:31:54
But then there's like augmentative aspects which it started assuming and it's like but that's a really big one in my opinion. So I think that's where it can be really revolutionary because you know just making people here better. It's amazing, but it's already being done and so I think the augmentative stuff is like, okay, is it possible?
02:32:12
That we're gonna have new capabilities as human beings with this type of technology. I think that is possible. And I think it's also kind of important differentiate like yeah where where where like what what are those types of new things new capabilities and yeah and in terms of tragedy, you know, I mean, That like yeah if you have an implant you don't you might not have as much control over your thoughts in your your interface with like the reality and with the internet like as you do on a computer you can stop using it if you want to or use a different platform.
02:32:52
And in that situation as well, I don't actually think there's a huge difference and we have to learn to like differentiate between that that that's that tragedy and that problem and the kind of mechanism that is happening on because right now we already have the tragedy of like you just Don't have privacy on the internet the these companies own everything you do and you don't even have a choice if you why use the platform you're handing off your personal life is kind of like if you wanted to buy a dishwasher or a washing machine your house or like plumbing you have to have a microphone and a camera in your house.
02:33:31
Just because that's how plumbing works and it's just that's pretty much what we're dealing with on the internet. That doesn't matter if you don't, you don't have to have an implant. You just want privacy and you don't have any control over your personal data. So, a lot of people will say like, oh neural link is going to prevent us from having controlled privacy like no that's already happening.
02:33:51
So that's been two cents, Allah , just one perspective here. I do think that's where it's heading and I mean there was a time where there were multiple subhuman species. I guess you can say derivatives from apes neanderthals and human savings are just the more commonly known ones, but I feel like we're heading towards a period where there's so many moral dilemmas.
02:34:21
As and various ways that are tribes can organize around, you know decisions as relating to neural link but also genetical engineering and all these advancements that we might have like, you know, just like today like there are there are tribes in Papua New Guinea that are completely disconnected from the rest of society then you have the Mormons who try to avoid tech in the you know, I guess the electrical sense and internet sense and but you know, if you pick up a tool, I mean one point, When humans first picked up their first stone tools and made those tools and harness them that that's it that was a form of tech so I feel like there's like a spectrum and we're just gonna continue along that spectrum and maybe have like a period similar to that period where there was various types of you know, almost species not just homo sapiens and where we would basically have some folks who are cyborgs with neuro-link implants and plantations and genetical genetic engineering upgrades and some who basically, You know are opposed to some of these concepts and might try to avoid it all together.
02:35:38
I do think in general it's a good idea to have diversity just as a species anything can happen but but I think it's like swimming upstream it's just gonna force the most people to opt into you know, whatever progress is kind of heading heading towards and ultimately the vast majority of probably have like society's gonna have technological implants if not in the next 20 30 years.
02:36:07
Maybe in the next 50 years. Whether we like it or not, but I do think yeah, it's gonna come with its tragedies but I do think we're gonna have we're probably gonna go through a period where there's like many what might feel like many different types of species because of our relationship with tech.
02:36:30
So um again on I guess like I also agree with with what apex for the most part so the revolutionary part for New Orleans for me is to miniaturization the fact that they are able to when it tries a device that small and can be linked to you know, the brain and their love toe designs cut has also certain some novelty because it can detect on a discrete level of micro columns, but the the tragedy part.
02:37:00
I think something like that really worries me and I don't know if this is actually mutual. This year by everybody that has gone through the mural implant is that the implants don't stay long like the electro types of degrade through time because the brain is actually highly corrosive environment it's a it's it's has like rich sodium in it it's like, you know in planting something you do an ocean it is not to say that the neural lane collect towards our metal but there is also the risk of you know, electrodes being raking and the the immune system kind of attacking the electrode.
02:37:40
Eventually reducing the signal's strength and also raising some kind of cytokine storm that would cause a severe damage to the brain so I would see the tragedy part and that but this is my opinion. I don't know if it's mutually shared by others or other scientists here in this group.
02:38:04
It's interesting the regular maintenance aspect of it. I wondered about what we were really worried about. I guess that the conversation we brought up was due to the topic at hand about, you know, the revolutionary or tragic aspect of it. Regarding the bioavailability of all of these in these implants and the issue that you just brought up regarding maintenance.
02:38:30
I know if you had actually studied this and actually gone through an experiment where you implanted something in a human being and you talked about discarding being minimal and it's not fairly insignificant. Repercussions from the damage to the brain are similar to existing brains to job applications, but I'm wondering about what you just brought up with if we're constantly upgrading our additions.
02:39:02
I don't want cybernetic conditions to come, but essentially that's what they are. What does that look like for when they break down? I guess I guess to try to form a thought or somewhere just a bit better. Earlier on we were talking about. Parameters that we could put in place to protect these things and what happens if you have a mouth or if you're dependent upon this tool to actually utilize something a computer or interface with something or not and the system breaks what sort of repercussions to that have and I was thinking at that time about a security option a hardwired kill switch that you could implement if in the event that something did malfunction you could simply turn all of these things off.
02:39:48
Has there been any discussion about implementing something like an actual physical or? As best you could get a physical kill switch for these implants to simply cut the contact it would have to the brain if they are implanted. So Pat so nerling is powered by you that you have to stick something behind your ear in order to power knurling.
02:40:11
It's not, it's not, there's no battery inside your brain.
02:40:19
That's actually.
02:40:26
Well, that's so I don't know if they changed it but that was what I saw in the original demo. I don't know if maybe that's not true for the monkeys but I imagine that they heard that for human beings it would be there. So the connection is severed then.
02:40:43
You simply turn if the device can't receive power because like imagine if you take this home you're going to want to you not want to you don't want to do surgery to replace your batteries so it makes sense that if you're not if you're not a monkey in a laboratory that you would you would have you would stick a device behind your ear and that's how you would power the device and if you if you needed to take it off you just you know, like in an emergency way you just pull the device off from behind your ear and what it's going to do is remove the power from the device.
02:41:17
Something tells me it would humans generally tend to to migrate towards Wi-Fi capability more than hardware though so I would assume that new way that at some point time they would they would try to figure out some sort of wireless capability or wireless functionality simply from a convenience play but that's really interesting my my my question it came out a bit weird but it was actually around the security around everything so we were talking about like the tragic consequences of having manipulation and it could even be the sense of or the case of people sending signals to these transmitters to.
02:41:52
Do cause and auditory hallucination or some sort of interlock like a man in the middle attack if you will inside the brain and I was thinking like the easiest solution to prevent that in the event that some signals were getting stimulated inside the transmitter which just be it like a kill switch someone to what we have on high security computers or cell phones now where you can simply press a button and it actually severs the connection between the cam or microphone etc and the the computer so that it can't be accessed in the event that It has been breached by it, you know a significant actor who can maybe turn it on when no one without notifying anybody or using the.
02:42:34
So that that was kind of where my train of thought was going with it and then I thought regarding the maintenance issue if these things are malfunctioning it would be it would make the most sense simply just be able to turn it off but it goes back to to that question about regular maintenance and everything like what happens if you have a significantly misfiring.
02:42:55
Something inside your head if I can throw a wrench and in the conversation here it's the idea that you that future attacks might be so sophisticated that you don't know that if there's anything wrong I and so you wouldn't have any there be you wouldn't have any reason to take to to hit a quick kill switch or to pull the the power and no one else around you would have any reason to either if you just think everything was fine.
02:43:35
So I'm probably gonna answer sub point because like you raced a very good question earlier and it's time for me to turn in any way so I just want to transfer that before I leave so there is a floating theory of that a memory can be encoded in the network itself instead of like simply by strengthening this synaptic connection between neurons and the theories called connect home, so it's gaining wide calculatority and there.
02:44:10
's been some computational neuroscience- like aspects trying to at least demonstrate that it's possible to encode some kind of memory at a higher dimension manifold and this was done by the blue grain project. I think it's in Switzerland so it's a very interesting idea to see that, you know, the network itself is capable of storing information instead of like I said, you know, by nearly strengthening it.
02:44:45
Synaptic connections in the brain. So yeah, I just wanted to mention it out there and this was in a great discussion. Thank you guys. Thanks for joining us. I had a question and you know Matt's offered to continue the room if anybody wants to continue but I had a question before dropping off or for Nick and anyone else who thinks they could sort of help provide some around this.
02:45:10
So, I mean, if you had mentioned obviously the work that led up to neuralink the precursor and you're your involvement. In it may be you might be able to comment on this but like what like specifically because a lot of folks when they think okay, you're linked commercialized something, you know, like they might overly simplify what that means, you know, it's not simply, you know, putting up a billboard and saying, hey, we want we have something we want to sell you.
02:45:39
There's likely a series of maybe smaller engineering problems or problems in general that they have to have gone through and solved to achieve economies of skill to achieve, you know, some of the efficiencies that they have. Could some of you guys maybe chime in on some of those recent things that they would have had to have solved in the last few years to get to where they are.
02:46:05
I know this sounds pretty ignorant because you've already mentioned that a lot of what they've done has been done in labs today. So what I'm really speaking to is, you know, just along the steps of like outside of other core things. I'm convinced they have.
02:46:30
Been invented anything new and I actually had that discussion offline earlier this morning. My question is more around like what are the specific problems that they did solve? Do you know that helped us get to the current technology that they have today? I imagine there were at least a handful of engineering or other types of problems could anyone comment on that.
02:46:57
I have an answer but I'll let anyone else who wants to go first, go first.
02:47:06
Okay I guess it's me So early income they created a microsurgery robot. It implants these microfibers with each one having like 16 electrodes. With such precision it does like really microscopic imaging, you know, it it's a it they're able to I have pictures on if you look at at the faceobok group self-aware networks or the faceobok group neurophysics these groups that I that I've been I'm running for 10 plus years and some instances.
02:47:36
There's pictures up there on the cover photos that show you the basically neuralink. What it looks like through the camera. They basically imaging the blood vessels that are really high resolution and then the robot is using them. Had a neural networking algorithm to to figure it to plot to identify areas where there aren't blood vessels so that it can pick areas that are a good targets for surgery and then a neurosurgeon is supposed to approve to say I agree with what the machine is saying these are excellent targets to implant electrodes and then the machine has to do it because because the actual surgery is too the the needles are too delicate for human hands to handle and and and the the it's the precision required is is is I believe it's too much for for the hands of a neurosur.
02:48:25
Geon so the robot has to do it and this this some this way like imagine so I don't know if you if people saw the movie Prometheus but then the movie was interesting because it's not the first movie to have this but it's it maybe the best reference it had the movie has a robot in it actually it has like a capsule machine that does surgery so at one point the character has an alien and her stomach or something.
02:48:48
I might be I might be messing up the story but she has something she wants to get out of her stomach and she goes to this surgery machine and it and it's only calibrated for for men, so she's really upset about that, but she said, That okay I want to remove this thing from my stomach and then the computer does the surgery automatically so the idea that we had that in the future we will have machines that do surgeries like a vending machine that does a surgery that that's that's sort of like a big component to mass producing medicine in a hundred, you know in the next hundred years it's basically to have the precision of surgery so well done by a computer that is better than a human being could do it and then you could eventually put it in a vending machine and and mass produce.
02:49:35
Surgeries alright, so that is an incredible innovation.
02:49:42
Makes comment my my name is shorts of a new religious so obviously I don't do research in this area but I I am familiar with technologies that I currently being used for clinical practices and and these modalities have been standardized and tested and they're pretty safe so well what you're telling me with the with the grids place grid of electrodes placed on the surface of the brain, this is something that is done currently and and you are right there are several types of machines or robots, if you want to call them.
02:50:17
They help people not only to place these grids in the surface of the brain but also to find out where you're gonna put them so this is used mostly for people who have complex type of epilepsies and where medication is not working and you want to look alive where the seizures are being triggered so you can maybe remove them through surgery and in these this technology is broadly used to obviously not in countryside hospitals, but in the major centers where you have, Prehensive epilepsy centered This type of procedure is performed.
02:50:57
And I am assuming that there is some type of collaboration between neural links and companies that already do these devices because it would be silly for one not to use this technology that's already available and try to reinvent the wheel, so I'm pretty sure that there will be some collaboration there.
02:51:15
The other aspect that I think it's really interesting to notice is that there is another company that produces this thing called stent road. Probably some of you in the audience have been. Seen this before and am familiar with it, but this is a technology.
02:51:34
But they are being planted in the brain in the blood vessels of the brain. They've done a few studies with shapes and I don't think they use monkeys. I think they used mostly shapes but they're starting to do this in humans as well. And the goal of this technology, obviously very different from neural link, but these electrodes are being placed in the motor cortex, so the part of the brain that controls movement.
02:51:58
And they're using these electrodes to control artificial limbs and people who have been victims of amputations or war. I've had mutilations and so forth. There isn't older technology that's already readily available but that requires neural surgery, but then shown is doing a similar thing minimally invasively. So, they just go the same way people have a catheter put in their hearts for when they have a heart attack.
02:52:25
It's a similar concept where the stint goes into a blood vessel in the brain and that's how the process starts and obviously everything is transmitted wirelessly. Now, just taking a step back when I think it was a mess who was talking about. Functioning Obviously we do have a lot of brains devices that are available and don't know if you guys know anything about deep brain stimulation but this is also technology that's used for modulating brain activity not for cognitive functions, like neural link is planning to do but for people for example who have injectable parking so disease or also people with epilepsy people with various types of neurological conditions.
02:53:08
So these are devices they're very safe they have they are made of material that are they don't do not cause an inflammatory reactions to the To the brain tissue as was mentioned before and I'm assuming that this is the type of material that also neural link is planning to use because again we don't you don't want to reinvent the wheel if there's something that's already very very safe to use.
02:53:30
Now, these are these devices the deep brain stimulation, they're probably the ones that are most used in the medical in the neurological community and what's interesting about these devices is that once implanted very deep in the brain there is a wire that comes externally and under the skin out of the skull and under the skin and ends.
02:53:53
In your. Chest under the skin in your chest and through there so there is a battery there that operates the machine or the device but also through there wirelessly with a with a something like a tablet or your iPhone you can control the amount of stimulation or or inhibition that you want to your to the structure of the brain that's causing trouble so all these technologies are there obviously enough for the for the for the purpose that's neural link is using but things that are safe things that are due can be used quite widely.
02:54:28
And and it's very doable so I want to believe that these guys from neuralink are talking to people who are manufacturers who are doing these non-devices just finally the main problem when these things stop functioning is that most of the studies that I've done for example one day in your life, you will need to have an MRIs kind of your brain and as you know, an MRI is I'm giant magnets so a lot of these devices they are MRI compatible provided that the battery is functioning now, we probably assume that if the battery is flat it is going to be safe as well, but all the studies were done with the Patients with full batteries so this is always a problem that's if someone has a device like this and they are needing an MRI for other reasons and the batteries flat nobody is going to go ahead and do it and once you have a device in blended in your brain the risk scar tissue the forms around it and get a getting rid of it is going to be very difficult if not impossible without causing any damage thank you, okay, can I love to respond to with George did you say Georgia Jorge?
02:55:34
You guys whatever whatever you prefer man call me anything cuz I got I love you and make me go I'd love for you to find George but just one question like the applications that George has a big like are you guys very familiar with some of the technologies referring to yeah, but thanks so much and have a great night take care.
02:55:59
Place. I'm heading out as soon as this has been fun, okay, thank you. Take care of the cashier. All right, so in fact, um, if any um, let me react positively if anyone else is leaving . Go ahead and say, so I'll just leave. I'll be quiet for a second.
02:56:20
Sounds good so um, in fact I have written about the interviewed doctor. David Petrino a PhD at Mount Sinai you okay? And right so Tom Oxley , he's an instructor and a director of innovation strategy for the Department of Neurosurgery at Mount Sinai and he specializes in vascular neurosurgery and interventional neurology . That's my article if you go to SVGN.io.
02:56:56
Um, and you type in the words dense road s t e n t r o d e you can read the read the article and includes the includes a recorded audio dialogue that I did a withstandro so since central is it does not require brain surgery it's amazing so you can inject it you can inject it into your arm.
02:57:20
And then you push it up through the through the blood vessels into the into, you know into the brain but but in the blood vessel somewhere above the the motor cortex is as it's how has already been said and then what happens as you withdraw the the device that that what happens is the the stent after it reaches it where it's supposed to be and you withdraw your your implantation device what happens is the standard expands inside the vein so that it basically it it's um, It basically presses outward against the walls of the vein so it it's it basically it allows the maximum amount of flow through through the vein it's not going on a block the vein at all and and then the electrodes are basically in the wall of that of that vein that's above your your motor cortex cortex just click your day and my cut this is the same technology that's used for unblocking a blood vessel that caused a myocardial infection a heart attack, for example, or if someone has a big stroke keep going, um, right so I didn't know that but that makes perfect sense but but it's like, If they did that and then they just okay, well we're not only working we're gonna put this in someone's brain to put open their blood vessels, but we could stick electrodes there at the same time that's that's that's the application.
02:58:47
You're learning. Or just as you speak as well, which is really fascinating for me, it might be hearing this correctly because so Michael yeah, he does enjoy it.
02:59:01
You said to unclog. The sort of applications like that and these guys have flipped it and used it to now insert wires is that is that correct can I tell you the story yeah you started because I was there when it stopped. Ped so Tom Oxley was a neurology trainee in the hospital where he used to work in 2011 and always a very fascinating clever guy so this is a technology that back then for neurology was being used for retrieving clots that were there were causing strokes in big blood vessels in the brain and this is a technology that's been available probably since 2014 or 15.
02:59:39
I would say no sorry 20 at 2005 2006. But it really became a a hype in the in the late 2000 of first decade of the 2010s and and this technology now is amazing so this is pretty much a stage which is a device that goes flesh into a clutch opens up and then when it opens up clot gets attached to it and they pull it out very sophisticated way of doing this so Tom was a he actually was a military doctor before he became a neurology training.
03:00:20
And he took a year off and he went to the US and one day he sent an email to the guy who was the one of the directors of DARPA and somehow he got a meeting to see this guy and he said now I have to come up with a clever idea.
03:00:34
And he had this he tells me he had this light bulb moment and he thought why don't we do this? I'll tell this guy that we're gonna use this for putting recording electrodes in the future. I use this technology to help you move artificial limbs.
03:00:52
So he was pretty much a combination of two technologies that were already. Available. The unblocking of the blood vessels plus the recording electrodes that I used for monitoring people with epilepsy. Keep going. Michael. I don't know that I have a whole lot more to say except. I just amazed him.
03:01:11
I shared the story you wanted. To add to that or but this is so what exactly did Neuralink do? Where did they step into this story? Like well this technology existed. I know when Dave was here earlier they were talking about a different type of technology that was being inserted in the brain and he had a picture of it on his profile earlier and so what exactly did nearly do with these things like that?
03:01:44
Um, Hey sorry I was having a little trouble with my mute button there. So, um, what I want to say, first of all is if you click on my profile right now and visit my Twitter and look at the latest thing that I shared not my pinned post but the latest post that I just shared that is a link to the story about centroid.
03:02:05
So that will be basically the best starting point to read is to read that story and and listen to the interview that I recorded. But so it was also brought.
03:02:23
You know, we do. I I'm kind of like I'm a little bit at the at the end of of what I can say about right now there's more in the article but I sort of like want to get back to to DBS and early that's where my this is right at the moment, maybe maybe go back to center later but um so deep so deep so when neurolink first, okay, Before you leave, um different it is it just for clarification arteries or vein?
03:02:56
So the distant gold. The original stage designs for retrieving clots was going into arteries but these arteries are of a certain caliber and this tense that had the electrodes was too big for it. So central to the technology is slightly different because it's going into big veins that you have in the surface of our brain.
03:03:21
So one of them goes from right in the midline from front to back on top of your other brain right underneath your skull and then there are two other ones that go at the back of your head to some. From the midline to the sides and then down your neck.
03:03:38
So so what I what I heard which is not not going to be as accurate as what he what which is said, but what I the the story that I wrote down was that this entrode is inserted via a catheter into the jugular vein and then maneuver it into the brain via via the vasculature and there's no there's no base basically the see I just wrote that there's no need to drill through the human skull the sensors are built until a self expanding stent that expands into the wall of the blood vessel.
03:04:06
And in time it's expected that cells will grow over the sensors and incorporate them into the tissue and then the signals would be transported out of the brain through the vein to a unit implanted in the chest and from there the signals from the brain would be sent to the computer.
03:04:21
I don't know if it's still up to date but the sunshard records rain activity in a way that's I guess similar to EEG or or ECOG which are like surgically implanted electrodes but the advantage of being inside the school. It is massive. I mean, that's just basically the field of ecog research, right?
03:04:41
And yeah, I'm just skimming my article to see if there's anything else here that I wrote down. But yeah, so check out check out you can check out pain pain medication reimagined with it was something that date. So David Trina was talking about the fact that he's a PhD but he was talking about the work by.
03:05:06
You said his name earlier. By Tom by Tom Oxley who is Tom J Oxley MD and PhD part of these cerebral vascular centers. Yeah. Anyway, so I mean, I think that it's just like I think that you know. So a lot of people this is interesting because well a lot of people who have been studying Copenhagen and effects on the brain have been like well we don't get how Covey 19 could get into the brain.
03:05:38
We don't understand why why it's affecting the brain like what what is it, how is it where are the, Are the the ACE2 receptors and well I guess maybe it was overlooked but the brain has a lot of blood vessels and it has a lymphatic system and throughout the blood vessels in the lymphatic system there's there's there's a there's a lot of whisk that, you know, there's a lot of cells that have what's called an endothelial lining and that endothelial lining has ace to receptors so I don't know why it's it's it's a big mystery, but but but yeah, it's seems pretty obvious.
03:06:20
Pleasure co-hosting our disruptors' done sessions. We had a lot of new faces today and lots of new speakers. A lot of our topics including this one today are sponsored by audience members like you. Also if you're interested in contributing or have any feedback please reach out to my selfie we're always looking to expand our group of disrupters.
03:06:40
We're building this great platform for you to share your thoughts and introduce yourself what you're working on and also finding like-minded amazing people that you can connect with and more depth offline. And I absorbed a lot of connections being made live as a discussion with running throughout amazing. We do this every Monday to Saturday 9 to 10 PM Eastern Time.
03:07:02
So everyone in the room right now hit that disruptor's icon at the top and hit that ball button. Bring a friend, invite people that you think would be a great addition and yeah, let's continue this discussion. Sorry for interrupting there. Yeah. Sounds great. Thank you Mayra. Okay sit well I wanted to add that so George also brought up DBS and so when the early link was first announced they made a lot of comparison to DBS and they pointed out that they're they're devices much smaller DBS the the implantation of DBS can can rupture a lot of micro micro vessels microblood vessels.
03:07:43
That's part of it's part of the I believe that's part of the reason that there's a long-term sort of scarring issue: the device is really big compared to an early link. The idea here is that. Going to cause far less scarring than you know. I don't know that the materials are that different with far less of them.
03:08:03
There has been some research which was sort of pre-narring which suggested that if you make your if you make the devices you put in inside the head made out of graphite that the immune system would not react to them and and there's also an airlink came out with one of their announcements was that they had implanted the devices into pigs and the pigs are you know, they like to bing their heads together, they're little they're the rock stars and so far the the, The then devices were not breaking so I guess they were thin and of it small enough and light enough that the bank the pig could being the head it's own head it gets other pigs heads and and it wasn't causing the these really light and small hairs to which contain which contained the electrodes to break and that's that's interesting because you know, the brain is basically.
03:08:57
Like it's really soft, it's almost like jelly or butter. It's a it's painful to think about how to delicate the rain is but um, but yeah, so the if you put something, you know heavy in there then and and it gets knocked around and then the potential for scarring like with the DBS the big DPS system seems pretty like a big deal and it just putting it in there as product could cause bleeding and and other problems.
03:09:23
In fact, like the first 10 years of brain surgeries, brain surgeries and in the history of medical neuroscience, this is True story. The first 10 years of brain surgery, not a single patient was surprised because they couldn't figure out for the long 10 years. They couldn't figure out how to cut into the brain without killing the patient because there are literally blood vessels everywhere everywhere.
03:09:46
It was just extremely ; it took 10 years for humanity to figure that out. So Michael, It sounds like there's a bit of material engineering work that was done in terms of the advancements they did with the actual material but how did they advance the technology so that they can minimize scarring and cutting into blood vessels?
03:10:11
Was that a bit of software engineering in terms of, you know, analyzing the topology of the brain before they did the insertions? What sort of stuff was going on there in terms of advancements that enough Georgia makeover can answer them. So without working out I can just answer that in a very general way that very very fast basically to make surgeries better unless let's traumatic on a patient you have to make them less less invasive less impactful you have to make the the surgical the surgery has to be small the cutting area has to be smaller you want to stick as little as possible in into the human body, you're human brain and you want to you want to do really good imaging ahead of time to figure out exactly where you need to cut and what you need to cut and within so an innocence the microsurgery robot that neuralink belts does two things what does it does really?
03:11:03
Hybrid solution microscopic imaging of the brain area where it's going to be implanted. So you can get a look at the blood vessels, but then two it also figures out where you know, what areas are the best for implanting the electrodes and then it implants them.
03:11:20
So that's basically the general idea and then of course, you know, there's a long history of material design and and and trying to find stuff that's at least offensive to me being. I think that answers a question without specifics. Yeah, I mean this is incredible this has been really really helpful folks who some of you that are contributing left but you know, I initially walked into this whole like when I started observing your link.
03:11:49
I it looked at you know in the realm of R&D and I had no idea that this there's this and this much prior work that has led to let up to their their their I guess contribution to the overalls, you know advancement of this. Of the stuff, you know, this is you know, you look at Tesla like there was an electric car that Elon had acquired.
03:12:18
Now, obviously there was a lot of work that had to be done to get to where it is today, but the concept of you know, the strategy and the concept of you know, starting with a luxury car working your way towards an economically feasible one that was baked into the initial vision.
03:12:39
I think of Tesla knows first started and in many ways like when I when I'm listening to the, Various domains and the on the prior applications that I've already been tested including the monkey example, including now the surgery that George is talking about here and you mica it's like yeah it's it's a credible to me that a lot of a lot of people don't know just how much work has led up to this moment and maybe partly due to Elon's popularity and and what you know recognition but and also their contribution to commercializing this I don't want to under you know value that but, It's it's I'm just shocked at how much prior work in invention and application and even understanding these use cases had already existed and had been discussed before neural showed up to the party and you know, it's really impressive stuff but just as now I'm blown away but all the work that you guys are have been involved in and thank you for for you know, sharing all this so I I think there's there could be a big perspective shift for everybody if you just think of this this is this is not Elon Musk's company.
03:13:54
Just just imagine that he had he he is he's the guy who's funding it that and now the reason I said that I say that because what I what what I'm getting at is like, you know there there was someone who basically you know it I forget his name but there was a neuroscientist who was the first to measure EEG signals and he he thought that you know, basically he read this letter where he heard that a woman had written about like her boyfriend.
03:14:29
She had the dream that her boyfriend fell off a horse and at the same time she wrote a letter her boyfriend had fallen off the horse and so he thought well well maybe our brains maybe there's maybe it's telepathy so he wanted to measure to to see if the he wanted to measure electrical signals in the brain to see if if he wanted to detect them and analyze them and he did and he was he was basically that like the grandfather beg.
03:14:58
Because he wanted to to see if telepathy was real and it turns out that he also he did he did really good work and it turns out that he was able to to his own disappointment realized that that EEG signals don't travel far enough there's not enough power in them that they could leave one person's brain and and and travel to another person's brain through the air, it's just not.
03:15:21
It's not it wouldn't work that way but you know that so but in a real sense EEG and also, you know, just measuring electrodes there's another guy who was was famous for he he was a famous he was a guy who a neuroscientist who came up with I'm sorry, I'm not being very good with names tonight, but but he basically was implanting electrodes into cats into their brains and he figured out that.
03:15:51
Neuro columns he was the first guy to say well there are columns have a function and and at some at some point the idea that different parts of your brain had different functions, what was unheard of and in human history and so he's like yeah, so your neuro column does something special it's different from what the rest of your brain is doing and it's organized it's somehow to organizing what you know, the cats reaction to to being petted or something like that right and it's organizing this this dismental activity and so so so that so so we so then we had lots of You know, we had many decades of people sticking electrodes and cats and mice and and rodents and monkeys and human beings and and we had the of course e-cog and we had the Utah ray which was mentioned earlier and DBS and and so so all of these technologies so if you look at who's hired by neuralink, you can see that that one of the one of the guy one of the guys who's hired was DJ co and he was the co-founder of neural dust and neural dust I wrote about neural dust too.
03:17:01
I just remembered so you could you could look up neurodust on my website that I linked earlier but um but neural dust I wrote about the other co-founder so I didn't know I didn't I missed the I guess I need to write another one about DJ co but I worried about his co-founder and then this was before an early sort of like review had their big reveal and and shared who was working there, but if you look at the people who worked there and there's been a lot of turnover but the people who worked over time so I think it was George that brought up he says, I hope that something like I hope.
03:17:36
That that they're talking to all the companies and in the scientists who have worked with DBS and and all and and and instant road and and the thing is for a fact many of the world's leading scientists are working at at an early link and have been working at an early link or previously worked at knurling.
03:17:56
I mean, it's been a long nearly has been around for years and inside and people have come and gone, but in great numbers, but the thing is like there is a, Vast amount of communication happening between the the leading minds and in neuroscience for a neurology and and you know cognitive cognitive neuroscience in it and in it's just so yeah, so to George is like question this it's not it's not only yes the ought to be doing it yes they have for a fact been talking to you know, the leading minds and the leading companies this has been happening they these people have been working that nearly so that's just what I wanted to say.
03:18:40
I think. I think it's worth mentioning as well that everything that you're implanting the body needs to go through approval from regulatory companies not companies entered is like the FDA in the US TGA here in Australia, so every country has that so I'm pretty sure that if you want to take shortcuts and make your your implants work quicker you probably use substances they have been approved to to being planted rather than again trying to invent something that might or might not work.
03:19:19
Like an escort, hey you mentioned something about the scarring regarding implants being a factor earlier on I think we had some some previous discussion around that earlier this evening and people were talking about how it's not going to be so critical but how in your opinion as a neuros neurosurgeon, like how critical is it or what is the impact and is there a difference between?
03:19:42
Age of a patient based on their newer plasticity and their ability to recover from that scarring. So obviously the the age is very important the younger the person the more likely that there will be effective in less damaging regeneration of brain tissue that's no doubt obviously one side once you reach a certain age there is virtually no no change that someone can have a 100% guarantee that their brain tissue is going to regenerate.
03:20:14
So what happens is that when you put a device like this, I think I might have missed you the terms card tissue. What happens is that some tissue from the body organizes around the structure? So for example, this states that we were mentioned before the stench and cardiac stains and kidney stains and everything has tended in the body these days.
03:20:40
Once you're put inside a blood vessel there is a lining inside the blood vessel that's called endothelium, which is a lining of cells so this filium grows on top of the stage and the stens get. Corporated into the vessel so if you if you if you are small enough to go into the vessel or hey you have a camera that's not enough and going to the vessel when you and you look at the vessel from inside you're not going to see the stains because it's been endotelized as we say.
03:21:09
So it's going to happen with these uh with these devices that probably there will be tissue that will grow on top of it and make it part of the body. So you're not going to be able to see the device in its entirety but the problem is that as someone mentioned before, if you want for some reason to get it out of your body, then you're going to create a lot of problems.
03:21:32
It's like it's easy for you to build something with concrete and put it on the floor or on the ground, but when you remove it a lot of dirt comes out with it. So the same concept would Happen in the brain but the brain is so delicate that's how any any additional or any damage that you do to your brain tissue it's going to cause trouble in particularly with these devices like pluralink, for example, you're you're stimulating the probably the most important part of your brain, which is you write your intelligence your your sense of of self-be and and and and I don't think that some that this is the type of device that we'll be able to come in and out very easily so once it's implanted or prior to implants in the decision needs to be made very carefully.
03:22:19
The person is to be aware that for whatever reason this should stay with them forever. Exactly, yeah, very very thank you very much but eventually I'm sorry one thing that mentioned to you is that um, you know, I'm a neuroradiologist and I and I see what it looks like after these after electrodes have been implanted and and also for epilepsy surgery trying to figure out where the seizure focuses electrons get implanted for like, you know, just a few days at a time and then taken out but, But these electrons and and also deep brain stimulation electrodes and I don't know the numbers but are much larger than the type of electrics that we're talking about here with your elite being the size of a human hair.
03:23:05
I mean the electrics that are put in for epilepsy surgery mapping. I'm always amazed when I see these and they go, you know, go going all the way across the brain and then they're taking out and don't leave scarring so I can't imagine today can I talk to the reason why there's no scarring is because they're there only for a limited time right so the when you do.
03:23:29
This monitoring stuff for epilepsy you keep the grids for what's two weeks three weeks and then the pain and then it comes out because these patients when they are being monitored for epilepsy surgery, they need to stay in hospital. Can't take the the grids with you home be monitored home and then come back so they usually in hospital for a certain number of days so they can have a so they can have a seizure captured and then by door the the most the biggest number of seizures capture so you can be sure that they are all coming from that full foci and then you can actually through surgery or whatever method you.
03:24:11
Right now and I didn't mean to imply that the electrodes do come out within a few days but still you slice you slice through brain you displace brain as these electrodes are being placed so your displacing brain tissue and they're much larger than the electrodes that neural link is going to use the size of a human hair.
03:24:38
I find it really fascinating. I had my father. Like a type personality high functioning individual like a director of cyber security for an energy company and he was in his mid-fifties and suffered he started having some some symptoms regarding his motor control and he went in and he was diagnosed with cavernoma and had to have some some fairly significant brain surgery to remove some some malignant tumors, so they're not cancerous but they was it was, you know, significant impact on primarily related to his motor control functions and, That was back in 2018 and he's been in rehabilitation since and trying to recover from it and he it took him years but it with his older age there was a significant impact on his state.
03:25:34
I guess if you will now his intelligence wasn't soon to be impacted thank goodness, so he's cognitive but it's it's the motor control and when you mentioned I guess when I was kind of probing around asking about the the damage the potential for damage for these, It was more or less like what happens when a person who's younger gets one and like you said okay when you come they might have to live those forever.
03:25:59
What happens if the technology, you know, sort of deep frogs and a more or better implant comes out 10 years 20 years down the road. You know, what could you apply a different one or would you have to take the chance of removing it and risking yeah I have a I have an answer to this and and I'm pretty sure that our radiologist colleague also knows all about this as well, but in medicine everything that's prosthetic or foreign to the body does not last forever so and just to give you an example for so if you have severe arthritis of your hip and you need a new hip.
03:26:39
We try to delay the operation for as long as possible because we know that even the most modern prosthesis they will last for what 15 years or so in your body and then you have to change it again, so um, I'm I am and also with a if someone has an aneurysm, which is a like.
03:26:58
Ster on the side of a blood vessel and they burst into the brain they can have either coils or or or a neurosurgical clip supplied to them for these many reasons and.
03:27:14
The coils do last for longer but they are much smaller structures, so it's going to be very interesting. How long and only time. How long the neural link is going to stay in the brain without causing any trouble and also what certain age would be the age where you came will be able to implement these devices into one's brain because.
03:27:38
I wouldn't want a adolescence or a teenager to have this equipment into their brains because we don't know what's going to happen in the longer term and not only that our brain obviously all the cells are there when we are born but the connections they keep on happening to you are in your mid twenties and so.
03:27:59
Structures of the brain, they only finish. When you are in your mid thirties, so you don't want to get anything that's going to interfere with this process if you don't know that this is going you don't know how it's going to be behave if this is called giving you a benefit and giving you better connections yes, but how are you going to be able to prove the fights probably possible, isn't it?
03:28:22
Was a neurologist. I'd have to ask right if you had a you know, the original use case of neural link is for people who are. Paralyzed right to try to give them some sort of communication point to the outside world if you had an adolescent patient. Significantly benefit from up from a life a lifestyle.
03:28:42
By implementing a neural link, what would you do? How would you get the benefits versus the trial backs of providing them this? Device even though what you just said, right you may significantly alter their neurological.
03:29:06
Someone who had brain damage and then had paralysis of their bodies said that what she's saying whenever they're whatever the current use case for her neuroling is now is right when I believe it's for something like paralysis or the brain.
03:29:23
She has someone who has a brain. It has already caused irreversible changes to the dead area of the brain and you want to try to compensate for their lack of function by implementing something that's artificial and it's going to work. I don't see any harm in that. I think that this medicine is all about risks and benefits: every medication, every vaccine, every procedure has risks.
03:29:47
And so I think that in this sort of situation where you have even. Proposals someone who has a damage to their brain tissue because of lack of oxygen when they're being delivered by their moms if someone is he's paralysis in one side of the body because of that then there's nothing to lose like totally agree with with the concept of implementing things to make their lives better and have a near normal life, but if it's something just to enhance.
03:30:15
You have now and they are considered normal. I think it's still a little bit risky to be honest with you. I think that in the future we will have that opportunity but I think that we are very far from that time and that time will have to prove to people with brain damage that this works prior to using this as a brain enhancer, that's my point of view.
03:30:40
And the situation that you mentioned if someone was paralyzed. And
03:30:53
Replacement then what would happen anyway is that brain tissue would wear away and it's called while learning degeneration because it's not getting stimulated. So in fact you would be enhanced by retaining that tissue and I think that happened perhaps with your father when the surgery would. They could that there was perhaps some because of how deep the cover number was there could have been injury to the white matter tracts in the motor area and perhaps that's what causing the some of his weakness, but in the case of neural leak again, the the electrodes are going to be so small the size of a human hair that they would not disrupt the white matter tracts and I think what were you saying earlier was whenever it is time to upgrade in 10 to 15 years, you're just going to have to leave the ones that are.
03:31:51
Already there and get new. In addition, so I want to I want to I want to share something that I want to share two points one is that you know, so so the white matter tracts that this there's two things happening one is you have a Lego dender sites.
03:32:07
That are covering those are cells that no cells are covering the axons of the body of neurons that are and and those run great lengths they in some cases they might run from your head to your toes right they're really long but but was interesting is that the the a Lego dendritic cells are or the that that the white waxy stuff that is covering the neurons it's not like it's just it's not like one oligodendrocyte in one neuron it's like they're going every different direction and you might have you might have a thousand.
03:32:42
Different a legendocytes touching one neuron it's it's it's a it's a little bit chaotic and so going back to George's point it's like imagine so imagine you've got a wire in there that's about the length of the human hair and then what happens is a cell if cell just randomly grows around that wire not now cell it in the cell is like a bubble it's basically a lipid bubble or a bubble of fat it's got it it really matters if that bubble is is has any punctures or holes in it that the that's a big issue for that cell and it's so if you just if you if it grows around a wire and you just pulled one.
03:33:17
Out no matter how small that wire is now there's a puncture in that bubble so you see the problem there and so if you look at in the case of you know, what is what a star scoby two doing to the body, well it's it's it's it's the the viruses is invading and it's causing these cells to burst and and I and maybe just having just just pulling that this the hair out of the cell might might cause the diverse I don't know exactly but that's just a picture.
03:33:44
I wanted to paint another thing for you. I think Travis did that because he believed that maybe. We were moving towards a world where everybody has these impossible devices, um, my thoughts are a little bit different in 20 years. I think that we'll have basically we'll have optical technology that will be integrated with basically like like open water technology but it'll be laser fast, it'll be like something like laser optical holographic technology that's integrated with ultrasound and and microwaves and it will not innovatively do effectively the same thing in early does without surgery outside your brain, you don't have to worry about.
03:34:23
It yeah. I can have two comments about this, okay number one. I agree. I'm.
03:34:38
You guys are familiar with the technology court transcranial magnetic stimulation, but this is a technology that's through the waves radio frequency, you can stimulate certain bytes of the brain and and and it's something that's been used for actually treatment of of depression because we do know some of the centers how they're caused or are responsible with the physio pathology of depression or how depression happens and we do know that if you stimulate these centers patients can have temporary relief of their depressed.
03:35:13
Ive. So train screen or magnetic stimulation is probably the closest we're getting to what you're talking about the problems that the. The two bridges of TMS are very wide and you can't have a stimulation of very very small.
03:35:36
Obviously I think that if we can get a device through laser beam or optical simulation, that'll be ideal, but I think that this subject is so. And sorry I don't I don't mean to to sound very unreasonable here or or to too much of a dreamer but I think it's probably going to be easier for us to think about computer singularity and sharing our brains with the hardware devices then having a stretch like neural link that's going to be able to control biologic body the way they are.
03:36:17
Thinking about it, I think the way singularities are evolving these days is so much faster than the way biological enhancement is evolving that I'm not entirely sure whether when near a link or the equivalence becomes available, we won't have something far superior for silicon bodies. I just want to say that I am very familiar with TMS but I do believe it's a it's over sold and it's not appropriate in all cases of depression especially cases where the depression was caused by a metabolic disorder because if the depression it basically the idea is, you know, your blasting the brain with a bunch of energy that's that is it is a you know, that there was a previous version of that and they're like 1960s I think in psychology where they're called electroconvulsive shock therapy which turned out to be really big disaster, there's a huge difference between a magnetic simulation it compared to electrical therapy.
03:37:13
It's far more it's far more targeted and it's it's like very precise in terms of the frequencies that is stimulating your brain with so it's it's it's different in many respects but the idea here is that if a person is suffering from depression because they're cells are not able to to process enough, you know, addendo triphosphate identity and triphosphate you know, or ATP if they're unable to make enough energy then then giving that person, you know, a TMS treatment is not gonna really help them in the long term that but it may actually make things work.
03:37:53
Express myself appropriately. I was just giving you an example . Like most things in medicine there's no such thing as one size fits, all right, so there is depression and depression there are tokens. Several times. Of presentation so yes. I didn't mean that this is a magical solution for depression sorry and and one other thing I am. I'm actually funded by NASA and we're using TMS to enhance cognition.
03:38:25
And so we're applying that over the frontal lobes and and and I think one of the biggest drawbacks about TMS is the depth that doesn't go deep enough to get to deeper structures like limbic structures in other areas. I think that's one of the biggest drawbacks of TMS and one other thing Micah is that is that electric compulsive therapy, it can be a disaster, but it can also be a lifesaver and it still isn't used today and in very severe cases of depression.
03:39:00
But in this case it can really be a lifesaver yeah, I mean, I am. I I I there's there's yeah, I can't disagree with you at all that um, there's and I think I think the idea behind behind DBS are deeper and stimulation is is sort of vaguely similar if you have someone who who's depression or or other, you know, PTSD isn't related to a metabolic disorder and they really just need more stimulation than then everything we just mentioned might might potentially including what wasn't mentioned TDCS those all might be helpful depending on what the patients needs are.
03:39:42
Changing what you want to say. Yeah, I was kind of curious, um. To Donna's point I think she mentioned this at the very end of one of her remarks where if you sort of have a version two of neuralink, you can just sort of like leave the version one in there and I guess just let it sit there dormant and then install the version two I'm wondering what the like upper bound is for like version updates in that manner at some point, do you sort of lock the, Pathways that you need to accurately detect like neural activity or is this something that can be done essentially like indefinitely and the other comment that I had was.
03:40:37
Is it possible if the goal is merely to move a cursor around a computer and click so very just like minimally viable output to a computer could one conceive of instead of installing a mural length or a standard type device in your brain could you use some other points in the nervous system like your fingertip or your you know, your your arms?
03:41:07
I'm gonna just have less long-term like upgrade ability risk. To sort of like monkeying with your brain and then the other. Idea that I had was, you know, we're talking about. I wondered to what extent. This technology could be used in the following way, like imagine a patient who is on life support they're on a ventilator and intubated, however, they still are not brain dead.
03:41:46
I wonder if you know being hooked up to a neural link and being able to use the internet is preferable to death and you can take this one step further where you know, maybe one day we had. Of you know extracorporeal brain life support like a brain bag and you know, maybe after you sort of live your corporal life, you kind of just get thrown to the brain bag and then with neural link you can kind of elevate productive second life kind of connected to the computer that I mean, we're in or about lockdown so essentially we are just essentially brain bags connected to our computers contributing.
03:42:33
Anyway, I was curious what everyone thought of those ideas. I think one of the important keys that as far as I understand when people try to develop these are quotation mind control technologies, what they're trying to do is improve speed of action and I think that if you like you said our moving the cursor just by thinking can we use some other parts of the body?
03:42:59
I think that if you are, you probably can but you want to go straight to the source of the information so you can make it as quick as possible and I think that if we have a good interface. Moving the cursor and you compare the the speed that are the velocity that it takes or time that it takes way to move the cursor just by thinking or by thinking and then using your hands I think that just thinking that you're moving the course cursor without or by bypassing your limbs is probably going to be much quicker and I think that this is probably the main goal of this technology it's going to be used for tasks that require quick response and precise response and and if you think about your hand if you get your brain, To get a small object obviously you will be somewhat precise but like I was mentioned before you're not going to be as precise as a robot for example and that's why neural link is developing these robots that are very precise for implementing these are these are devices in one's brain, so I think that that's the goal and I think they're trying to do it some of the way we would be sort of futile any it would defeat the purpose now coming to the last point of your question.
03:44:31
And here a stroke never had any never knew that he had any respecters and he had a stroke in a part of the brain called the brain stem in the transition between a part called the midbrain and the ponds and that's a part where all the connections that go to your brain and come out of your brain travel through.
03:44:50
So it was as if someone cuts with a knife that parts the communicates your brain to your body and this guy had what we call any probably heard about this before what we call locked in syndrome. So he could only move his eyes up and down and blink and there was all he could do.
03:45:07
And this is a very clever person who's a surgeon, highly active and very fit. So if you asked me ten years ago, what would you offer this patient? I would say look this isn't there's no quality of life there. I think we should probably just offer comfort.
03:45:25
Measures and let him die Let nature takes it scores But now I am at a point where I still don't have I still don't think we have a solution for the problem, but when I was discussing this with the patient with and with the family, I was thinking should I offer these guys to keep in my life for five years hoping that's neural link or whatever is available is going to help this guy in the future or should it just pay the penalty for not being born in the right time in the same place and let him die knowing that at any time we can have a technology that will be very helpful for him.
03:45:59
And at the end of the day, the guy was a very proud surgeon and he decided with his family that he wanted to die and we actually let him die. But what a waste of brain tissue letting someone die just because they have no communication with their bodies knowing that their minds are intact and their cognitive abilities are fine.
03:46:18
Please help us make a, H, no, I understand. Interface has higher throughputs and would be more performance than. Something that would replicate the signal of moving your hand moving your lip but isn't that what we have now? I mean like when you type on a keyboard those are essentially neural activity that's being fired through your arm.
03:46:47
So if the goal is not to create like the super interface, if the goal is just to have a quality of life that's commemorated with that is epsilon better than dying. That is, Closer to what we have now. I mean like with that less risky in terms of what if able to extract or having the other consequences of putting something in your brain.
03:47:17
I mean the downside there's just like an imitator scenario rather than you have to be left. I suspect that we. Sorry George Jonah.
03:47:36
I suspect that if we can actually interface directly with our brain, it's not just going to be an incremental optimization from just typing on a phone. I think there's going to be opportunities to evolve. You know, language potentially anticipates someone who's trying to communicate a certain thing. It might even potentially lead to helping them string the right words together.
03:48:14
I suspect that there's probably some advance there but Georgia forgive mechanism but that's certainly. The patient that you just referenced it's just shocking well, this is just again, this is just one example of so many stories and.
03:48:40
Parents have been. And look there's always hope that what they technology will be able but I was thinking also that maybe an interface that might be able to control a an excellent skeleton that supports these guys body and at least give them some opportunity to walk around and do medical things that would make his life a little bit less boring than just sitting on or lying on in bed and watching Netflix all day long, you can only watch so much Netflix, right?
03:49:08
Yeah, yeah, so skeleton I think about that but that's the other aspect of it as well. I was gonna say this one thing you guys are all familiar with like how Tesla started with luxury cars trying to, you know, demonstrate that there was a market for this and and eventually used that to fund and you know create commies or skill for the model three the folks who are.
03:49:36
Likely to benefit from this like the the patient that you described if they can get experience creates if they digging to get advantages that clearly you know leapfrog ahead of us and you know, that's I think where the tipping point will occur where we see them having benefits they were once paralyzed but now they're communicating and you know being productive in a way that's far more efficient than we currently are without neural link.
03:50:11
I think that would be the tipping point that could occur in the coming decade and that could be where others would want to I think be called medical tourists. I think Michael was talking about earlier but medical tourism is the idea of going around. You know, to places where perhaps you can get your own years like implanted, that sort of thing.
03:50:35
I definitely if you think just think about hunter gatherers like you know, we today like, you know, I'm sure back then if you were a certain size a certain physique a certain strength, that was you know, an advantage for survival. Today, you know, like that's not the case and I can only imagine that how some you know, just in general technology and other advancements allow for many folks who perhaps.
03:51:07
And we're not biologically the most suited for survival 5,000 years ago today. I think you know because of technology because of all these announcements we were all part of this sort of modern civilization and I think perhaps neural link could do something just similar by the beating the charge with folks who are paralyzed and other types of issues, you're just degenerative issues giving them an edge over the, Rest of civilization and then causing that to be the tipping point for wanting to pick up these tools as well.
03:51:50
I think there's a couple of big things to think about, maybe haven't been mentioned. I wasn't here from the very beginning but um. One of them was. Oh gosh. I wish I wasn't drawing a blank right now so so one of them was connecting your brain to okay, so so when it's so back in 2012, I opened up a business in San Francisco that was a biofeedback neurofeedback business alternative therapy and the goal was to use EEG to drive lights and sound to your to you're too goggles that you would wear with your to LED glasses and into.
03:52:37
And headphones and so you'd be immersed your brain would be immersed in light and sound patterns that had that word being driven by your own eg patterns and the idea here was in my experience of it with it was that by listening to by by feeding these signals that your brain was your brainwaves were causing changes to back into your brain through your eyes and ears that you might begin to sense your your own brain activities to and also since how your thoughts and feelings were becoming brainwaves and were then changing.
03:53:12
This the license sound which was then so basically that the my my hope was at time that this would allow me to understand your greater extent basically to become self-aware of my brain activity of my brain waves and my and how my thoughts and feelings were brain waves and you know, I know a lot more now about the brain than I knew back then and then there's um, there's huge.
03:53:44
Sort of um problems with. Problems with the. Maybe it doesn't won't help that much to go into it, but but but the basic idea is that maybe with the greater resolution and definition of neural link if you could send that activity back into your own brain.
03:54:10
Maybe you would have a much higher resolution understanding of your own brain like sort of like in the like the way meditators so so there's there's there's all these um, you know stories about you know, I guess the big story is is but is sit harness at heart attack Guatemas at sat underneath the tree just for a very long time and he so and then he realized he was one with the universe and and then and then the Buddha and the Buddha right it's the Buddha story but but there's all these stories like well if you become somehow more aware of your thoughts and feelings and You could have transformational growth you could have happiness you have peace you could have you could have more control over your your own choices and you can make better choices and and you'd have enlightenment and you would you would and also like what if what if we could mass produce enlightenment for people become very in touch with themselves, very self-aware and very at peace they find very they like you like like imagine that think about all the unhappy people in the world well, well what if you could this create a device, okay, well, hey, you're unhappy but there's a there's a way try this device in a short order you'll be.
03:55:20
Come very you'll become enlightened you you'll begin great insight into how your mind works and how you're doing the things that are not working for you and how you can do things that do work for you that's one thing so the other thing is artificial intelligence, so when nearer link was first discussed I they there were two two ideas sort of seemed to emerge one is telepathy and one was artificial intelligence selected yeah if you can send if you can connect your your brain signals to a computer and then send them to someone else that's sort of like an artist's telepathy an artificial intelligence is a different.
03:55:51
It's it's a it depends on it I mean it it's it's it's a so I had this dream one time that I I was the dream is interesting cut because it's like one possibility. I'd have a million positive, you know a lot of different possibilities, but I in this dream I went into a cafeteria and I sat down and I though the accepted this chip in the chip was was a little AI and I could put it either on my waist or my neck and the idea is that it would tap into my nervous system and begin sending and receiving messages too much to my brain and and then I got up out of my body now.
03:56:25
I was based. On a lucid dream the computer induced the lucid dream the AI was managing my brain so that it looked like I looked at so it looked managing my brain on my body so that it looked as if I was sitting in a cafe and a normal way drinking a cup of coffee and just sitting there but then in the lucid dream that's computer created for me sort of like virtual reality, but it's like it's like an induced dream I was but and I was were aware was the dream and I was where the computer was creating the dream I got up and walked out of the cafeteria.
03:56:57
And and then I began to play sort of like an augmented reality game where I was in the world but now there was like there was angels coming out of the sky and they had like axes and swords and and and medieval weapons and I had to fight them and it was and this was this was what I one of the things that I imagined the future would be like in the stream, so I just wanted to share that and maybe with AI and VR.
03:57:21
I'll just teach you a very cool experiment so if you wanna be somewhat aware activity is like so obviously the brain electrical activity were at works in frequencies hurts, so one day try if if there's a bright light or or even I think the best way of doing this is a light down outdoors on a sunny day and that close your eyes, but make sure you focus on your vision and you see like a an inverse ripple if, That's the circles coming from the periphery to the center of your vision and their circle after circle after circle that is the frequency of your occipital lobes or the sequence or the frequency of discharge of electrical activities from your occipital lobes, which is where your vision is controlled and this is one of the modalities that people or this is the one that this is one of the ways that people use for starting their meditation just by looking at these circles and trying to disconnect from the outer world and going to their inner world if you do meditation try doing.
03:58:31
How cool that is.
03:58:35
So our partial baby and our hairs are one either. Under all just neuroscientists when I have some inside of this but my understanding is that the current like the deadliest that demo monkey or whatever that neural link had is essentially using the neural link for like brain output right there were controlling the pong game or whatever with their mind but their input was still sort of, you know, their eyes does neuralink based on what we know the current technology have a solution for the input or is the input.
03:59:15
Sort of banking on sort of like a prosthetic eye or prosthetic ear technology or is this not relevant is output essentially the is there a reason why output is the only method that is being focused on right now. I can answer that so so neuralink has read and write capabilities they these electrodes that can read your brain can also you can also transmit electrical stimulation through them and and so the with their talking about is creating closely therapies where you you send some, Emulsion and then you you read to see what the stimulus did and then you send some more stimulus and you read again to see if the stimulus did then you figure out if the therapy is helping or not helping that's the that's an international that's most important simple way to explain a closed loop therapy, but but but yeah, but in addition to that there's nothing that stops scientists then who's using a neuralink to to also use the eyes and ears in addition to the electrical stimulation provided by the device.
04:00:17
Right, so that's used for therapy but I'm saying that if I just wanted to like to project an image to my brain without using my eyes or ears is that something that is. Like certainly that wouldn't be in the category of the therapy right? That's beyond our technology now beyond our consent.
04:00:36
Ration. No, I don't.
04:00:41
Psychedelic drugs because it's projecting an image into your brain is not that simple. The pathway is so complex that having a structure that's by not using your vision is going to be able to stimulate so many different parts of your brain to trigger that image. I think sorry never say never but I think it's virtually impossible.
04:01:06
You would have to be something super sophisticated and and in unbelievably a complex but I think yeah, that's that's when psychedelic drugs can can help you with that and there are lots of places and studies that are showing that you can do this in controlled environments are very safely and the experience that people get our mind blowing and they help for number of medical conditions and even just to improve your performance improve your your levels of energy and healthy people are doing it because it's it works.
04:01:41
Probably that's our solution to the brain, the computer interface really. We're trying to use the computer. It's never but I guess what I'm saying? I don't think we have that sort of solution but with that just sort of be then an artificial eye that is like you just plug your eye into a computer and it's just sort of projects the image through your iPathway or maybe an artificial year is easier to create I guess we have these enter concludes already, but so I so I have reasons to disagree, um, if I can track so so I mean, look, if you consider David Eagleman's work.
04:02:16
He you know and in the work that he talks about done by others they they you know for example I brought this up earlier that they can plug a camera into an electrode to strip that they can put on your back or your tongue and then your brain can work out with the images.
04:02:31
They can also take an eyeball out of a rat and plug it into the auditory cortex of the rat and your brain can and they're the rat's brain can begin to see that way. So what that tells us is that it is possible to make the brain see through electrical stimulation, whether you stimulate the brain or stimulate the skin.
04:02:51
Or skim it stimulating the tongue; it's possible to make the brain see things. In addition our TMS expert left, but you know, she said she had also experienced that if you assume at the brain some sometimes if you made the brain with TMS people see a light that's not there.
04:03:07
All of those points to the fact that within it with brain simulation, it is possible to make people see things. But I don't. I don't know if I'm okay, I'll pause. Well, I was just gonna follow up and down what here was saying about psychotropic drugs or psychedelic drugs.
04:03:25
My understanding of how LSD is actually affects the brain is the similar to synesthesia rate where I believe that's the right term where it's essentially remapping the neural connection between the different senses and that's why people see they'll taste color or they will feel sound or something like that, they're neural pathways are basically not interchangeable but can be remapped.
04:03:53
In. Side the brain and to your point Mika vote having this sort of interoperability between the different sensory inputs. I think that there is a lot of research to still be done to really understand the availability of these neural pathways but I think that to sort of answer KKGs or KKK's question.
04:04:16
I think there is a good opportunity to reimagine the input interface and how it correlates to the neural. Gia provides the output to control a separate function like if you can imagine maybe if the visual reflector isn't available maybe there's a sensory one maybe there's a hearing one where you can actually control the neural link using some other sort of sensory function rather than simply relying on a singular one.
04:04:51
Sorry guys. I'll just quickly have to leave. It was great talking to you and I hope we can catch up some of the time. Thank you. Great conversation already, thank you very much for your participation. I think I need to go to bed, so I'm gonna head out to thank you good night.
04:05:09
Thanks again mika again, it was great to hear your perspective on everything. Thank you so much.
04:05:16
Yeah guys, it's been really great. I think it's there.