How to Upload a Mind (In Three Not-So-Easy Steps)

Подписаться 257 тыс.

Просмотров 150 тыс.

50% 1

Welcome to our tutorial on mind uploading. Follow all the steps, and you should be good to go.
▀▀▀▀▀▀▀▀▀PATREON, MEMBERSHIP, KO-FI▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🟠 Patreon: / rationalanimations
🟢Merch: crowdmade.com/collections/rat...
🔵 Channel membership: / @rationalanimations
🟤 Ko-fi, for one-time and recurring donations: ko-fi.com/rationalanimations
▀▀▀▀▀▀▀▀▀SOCIAL & DISCORD▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
Discord: / discord
Reddit: / rationalanimations
Twitter: / rationalanimat1
▀▀▀▀▀▀▀▀▀SOURCES & READINGS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
Turing, A.M. (1937), On Computable Numbers, with an Application to the Entscheidungsproblem. Proceedings of the London Mathematical Society, s2-42: 230-265. doi.org/10.1112/plms/s2-42.1.230
Sandberg, A. & Bostrom, N. (2008): Whole Brain Emulation: A Roadmap, Technical Report 2008‐3, Future of Humanity Institute, Oxford University www.fhi.ox.ac.uk/brain-emulat...
Francesco Randi, Anuj K Sharma, Sophie Dvali, and Andrew M Leifer (2022): Neural signal propagation atlas of C. elegans, arXiv:2208.04790 [q-bio.NC]
Rafi Letzter, “After Break with MIT, Nectome clarifies it has no immediate plans to upload brains”
www.livescience.com/62212-nec...
Eth, D., Foust, J., & Whale, B. (2013). The Prospects of Whole Brain Emulation within the next Half-Century. Journal of Artificial General Intelligence, 4(3) 130-152. DOI: 10.2478/jagi-2013-0008
“Dendritic computations captured by an effective point neuron model”, Songting Li et. al. 2019 doi.org/10.1073/pnas.1904463116
NVIDIA ADA GPU ARCHITECTURE, images.nvidia.com/aem-dam/Sol...
Beniaguev, D., Segev, I., & London, M. (2021). Single cortical neurons as deep artificial neural networks. Neuron, 109(17), 2727-2739.e3. doi.org/10.1016/j.neuron.2021...
Joseph Carlsmith, 2020. “How Much Computational Power Does It Take to Match the Human Brain?”
www.openphilanthropy.org/rese...
Top supercomputers 2022: www.top500.org/lists/top500/2...
Projections of top supercomputing power: www.top500.org/statistics/per...
▀▀▀▀▀▀▀▀▀PATRONS & MEMBERS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
Shrimant
RMR
Kristin Lindquist
Nathan Metzger
Monadologist
Glenn Tarigan
NMS
James Babcock
Colin Ricardo
Long Hoang
Tor Barstad
Gayman Crothers
Stuart Alldritt
Ville IkÃ¤lÃ¤inen
Chris Painter
Juan Benet
Falcon Scientist
Jeff
Christian Loomis
Tomarty
Edward Yu
Ahmed Elsayyad
Chad M Jones
Emmanuel Fredenrich
Honyopenyoko
Neal Strobl
bparro
Danealor
Craig Falls
Vincent Weisser
Alex Hall
Ivan Bachcin
joe39504589
Klemen Slavic
Scott Alexander
noggieB
Dawson
John Slape
Gabriel Ledung
Jeroen De Dauw
Craig Ludington
Jacob Van Buren
Superslowmojoe
Michael Zimmermann
Nathan Fish
Bleys Goodson
Ducky
Bryan Egan
Matt Parlmer
Tim Duffy
rictic
marverati
Luke Freeman
Dan Wahl
Alcher Black
Rey Carroll
Harold Godsoe
William Clelland
ronvil
AWyattLife
codeadict
Lazy Scholar
Torstein Haldorsen
Supreme Reader
MichaÅ‚ ZieliÅ„ski
▀▀▀▀▀▀▀CREDITS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
Producer: :3
Writer: Allen Liu
Line producer: Kristy Steffens
Production managers:
Jay McMichen
Grey Colson
Animation Director: Evan Streb
QA Lead: Lara Robinowitz
Grey Colson
Ethan DeBoer
Damon Edgson
Gabriel Diaz
Jodi Kuchenbecker
Jay McMichen
Skylar O'Brien
Vaughn Oeth
Lara Robinowitz
Background Art: Olivia Wang
Compositing:
Grey Colson
Patrick O' Callaghan
Narrator: Robert Miles
VO Editor: Tony DiPiazza
Sound Design and Music: Epic Mountain
▀▀▀▀▀▀▀CHAPTERS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
Chapters:
0:00 - The masterplan
1:30 - Scanning
4:25 - Interpretation
6:25 - Running the emulation
8:38 - Which one first?

Наука

Опубликовано:

12 май 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 796

@RationalAnimations 6 месяцев назад

We hope this new tutorial will serve you at least as well as our previous one about how to take over the universe. You're welcome. 🟠 Patreon: www.patreon.com/rationalanimations 🔵 Channel membership: ru-vid.com/show-UCgqt1RE0k0MIr0LoyJRy2lgjoin 🟤 Ko-fi, for one-time and recurring support: ko-fi.com/rationalanimations

@IcePhoenixMusician 6 месяцев назад

This emulation sounds like it might just be catastrophic because it’s simulating a brain… a person, but they aren’t getting any real stimuli. You’ll have to simulate a universe for them to live in too, or you doom them to insanity, if white torture results are anything to go by.

@ChemEDan 6 месяцев назад

A viral vector penetrates neurons, and upon uncoating, generates a random DNA sequence which is likely to be unique to the uncoating event by virtue of its size. An additional, non-random sequence is added to characterize the type of cell that was penetrated. These sequences are incorporated into proviruses and actively transported to each synapse of the neuron via its axons. A non-random sequence is added to characterize the type of synapse a provirus arrives at. The proviruses mature and release into the synaptic cleft as daughter particles; however, some are intentionally left bound to the cell membrane (as with a neuraminidase inhibitor) in case they get there first. When two daughter particles with dissimilar uncoating sequences and dissimilar synapse sequences meet, they fuse, ligate, and generate another random sequence for indexing purposes, and a final non-random sequence that indicates a successful completion. In this state the daughter particles are able to unbind and chemically ordain themselves for excretion by the kidneys. Then the patient metaphorically pisses their brain into a little plastic cup. Sequences are amplified by PCR using the success sequence as a primer. The sequences are read in a massively parallel fashion, but instead of 1-dimensional sequence overlap, an algorithm identifies individual neurons and their fully characterized relationships with others. This is possible because multiply infected cells will be represented by multiple unique uncoating sequences with identical connections. With the help of a reference connectome (analogous to a reference genome sequence), which may take the form of a neural net itself, inferences are made to render the uploaded mind operational. Hopefully. Now plug that stinker in. C. elegans would be a good proof of concept with demonstrated reproducibility, but I suspect modern sequencing and computation could take on a mouse. I guesstimate the human mind to be on the order of exabytes, so costly but perhaps possible. People might need to merge into a single entity for economic reasons, which raises all sorts of interesting questions. But delivering a sufficient number of viral vectors without killing the host might pose a larger problem, as could elimination through urine. Maybe a head can be kept alive under anesthesia, with artificial respiration and circulation, until the brain is ready to be extracted and homogenized. **Hannibal licks spoon** What do y'all think?

@sageinit 6 месяцев назад

Oh great youtube stuck my comment into the mod queue because I edited it to fix something, and/or due to the number of links to sources contained within it

@sageinit 6 месяцев назад

Let's try that again:

@sageinit 6 месяцев назад

I hope my comment now has actual visibility and ain't stuck in the mod queue due to the links.

@Deltexterity 6 месяцев назад

i'd just like to point out how phenomenal the soundtrack for this video is, huge props to the composers!

@smitchered 6 месяцев назад

Epic Mountain! :D

@Deltexterity 6 месяцев назад

@@smitchered oh i didn't even realize, i love epic mountain! been listening to them for years

@smitchered 6 месяцев назад

For some reason you they only credit the music all the way down after the patron list. I have spoken with the manager.@@Deltexterity

@pollo_frito22 6 месяцев назад

It was so immersive I didn't even realize it was there

@FarhadHakimov 6 месяцев назад

I can't help but hear Hotline Miami's soundtrack in this one. In fact, I'm currently watching the second half of the video with Hydrogen playing; and how strange it feels to have my brain find the idea of emulating a human brain and Hotline Miami perfectly making sense together.

@heavencanceller1863 6 месяцев назад

It's also important to note that even with computational speed limits, we might not have to run the emulation in real time

@jmoney4695 6 месяцев назад

Now that you mention it - it seems likely to me that the first whole brain emulation attempts will occur much slower than real time, slowly speeding up as parallelization and specialization improves until the brain emulation is running much faster than real time. Just imagining the first person spending an eternity in slow motion. Or maybe it would seem “normal”, because their perception would also be just as slow.

@dylanherrera5395 6 месяцев назад

@@jmoney4695 it 100% WOULD feel normal to them

@ludicrouS_406 6 месяцев назад

@@jmoney4695if the emulation would be slower than realtime, then everything would be sped-up to it, since the time between complete processes would be larger than the one of real brains, yet it would feel like nothing has changed. we would see them act in slow motion, and the emulation would perceive us as being sped-up. in fact, slowing the emulation time would make "time travel" towards the future realistically feasable (assuming the hardware won't break) as they could make 1000 years seem like a mere second to themselves.

@billiam6398 6 месяцев назад

⁠⁠@@ludicrouS_406The reverse is intriguing too, with more powerful hardware able to emulate a mind significantly faster than “real time”. reminds me of marvin the paranoid android. i imagine an emulation of a brain running so much faster than real time might suffer from something similar to marvin. imagine all of existence around you going hundreds of thousands or millions of times slower than it should. the time every task you could possibly perform takes pales in comparison to the endless expanses of time with nothing to do but wait for necessary feedback or a new goal. while conversing with an ordinary human, they would seem to pause for weeks between each response, months or years if they take time to think something through. at 100,000x emulation speed, an 8 hour down time would feel like 91 years, with literally nothing to do but wait.

@ludicrouS_406 6 месяцев назад

@@billiam6398 sounds hellish

@NicolasBPierron 6 месяцев назад

One problem might be that we over estimate what the brain is responsible for, and neglect the fact that it interacts with a body which also sends and interprets signals. One interesting example comes from pianists who are able to play a lot of notes compared to the limited bandwidth between the brain and hands, which suggests some form of co-learn compression, which would also have to be scanned / interpreted / emulated.

@bowfuz 6 месяцев назад

Replicas is some Keanu Reeves movie that's responsible for my becoming aware of this co-cognitive idea, if you haven't seen it I super super recommend it.

@dayvancowboi9135 6 месяцев назад

that's interesting, but paralyzed people exist and have rich interesting lived experiences, so clearly not having full knowledge of the body's system isn't esential to this

@bowfuz 6 месяцев назад

@@dayvancowboi9135 well of course, but if we're going to duplicate a brain and put it in a simulation, it's going to be jarred and will need some amount of physical therapy if there's any significant difference to its somatosensory experience.

@dangerfly 6 месяцев назад

Sean Carroll has a relevant podcast on Embodied Cognition.

@jaylewis9876 6 месяцев назад

It might turn out that makes it easier. Once one pianists finger memory has been captured that becomes a sub module for other pianists. Amateur piano players may notice an immediate improvement but won’t suddenly be masters since they would still be missing the nuances. Like giving a race car to a normal driver allows them to go faster but they still have to know when to break and accelerate to get the results

@Woodledude 6 месяцев назад

I think the Bobiverse series was a piece of science fiction that really drives home the fact that just *running* the emulation isn't enough to sustain a human mind. The task is made more complex by the fact that the brain requires at least some inputs at its own run speed - Putting a human in a dark, soundless box is considered not generally a fun time for most people. Much more so if they're divorced from even their own sense of touch, smell, taste, proprioception, thermoperception... Not having a body would fundamentally change how a human brain operates, and may require either fundamental architectural changes - Or the simpler solution, a simulated environment (and body) to replace those sensations sufficiently. Yeah, game design may actually be a crucial step in this problem. Weird to think about.

@michaelsbeverly 6 месяцев назад

It seems without feelings and hormones and etc., as you've noted, a mind would go insane. Just "thinking" is only a part of what it means to be human.

@trolly4233 6 месяцев назад

Poor Henry...

@Woodledude 6 месяцев назад

@@michaelsbeverly Yes, exactly. It is critical to make a particular distinction here, of course - The distinction that we are SPECIFICALLY talking about biological/anatomical hunans, and minds derived thereof. I firmly believe there are other possible kinds of intelligent minds, and presumably these would have different needs - Indeed, arguably a digitized human mind, even perfectly emulated, has distinctly different needs than a normal human, even if they stem from solving similar problems in the underlying structure. I think one of the fundamental takeaways we *can* get from this, though, is that intelligence requires stimulation, in the same ways that an organism requires food. Plants can make their own food after a fashion, and perhaps some sapient beings could make their own required stimulation, but this ability is not guaranteed - Certainly not in humans. We have a duty to ensure that an emulated mind has all the stimulation, or tools to create stimulation, that it needs. I suppose another way to put it is this - As far as humans go, their bodies are most certainly part of their mind, and removing the body effectively removes part of the mind. It is the same for the environment - It may sound strange, but your environment can similarly be considered part of your mind. Thus, to emulate a mind, we need to emulate the whole mind - Which includes, if not necessarily *THE* body, at least *A* body. Ditto for the environment. It doesn't even have to be a physical environment, just an environment that provides similar levels of stimulation, something to latch onto and provide some mental grounding.

@AuntBibby 6 месяцев назад

i was thinkin about this too!!! if a virtual brain woke up in a virtual naked body whose skin wasnt touching the air and the floor properly, (as in it was clipping thru the floor and the air was clipping inside the skin), it might feel so incredibly uncomfortable that the virtual brain would want to die!!! which means either the environmental simulation has to be absolutely perfect??? or the virtual brain has to be deeply sedated???

@Woodledude 6 месяцев назад

@@AuntBibby It's a very interesting thing to think about, isn't it? As a game designer, I can tell you that the sensations don't need to correspond 1 to 1 with what we would physically expect. They just need to be present with sufficient nuance. I imagine that one way the eventuality you described might be resolved is as follows - The material of the floor is effectively a "ghost jello", and the sensation of "clipping" would present the body with a sensation as if that material was pressing in gently around the foot, or possibly non-violently permeating it. So it would sort of feel like swimming through wood, or concrete, or metal, or whichever floor surface you're on, except it also gives you stable purchase to stand upright. Pain would actually be an unecessary/elective sensation, most likely, seeing as its purpose is to alert you to bodily damage that shouldn't be happening in a controlled simulation. It would actually be more difficult to implement this kind of thing with the physical accuracy you're describing - As I've eluded to, this is more the realm of game design, where effective cost-saving shortcuts reign supreme. It would be critical to present a mind with a non-traumatic environment in the first iterations of these simulations - Any competent designer would deliberately exclude pain sensations until they could be safely tested and debugged. I hope this soothes some of your fears! There's always the possibility of malpractice or negligence with this kind of thing, of course, but hopefully these kinds of discussions help make people aware of how this kind of thing should be thought about, and what kinds of questions to ask.

@px43 6 месяцев назад

The TV show "Pantheon" is an excellent and criminally underrated series that explores the ethics of mind uploads.

@lilithkiwi 5 месяцев назад

clicked on this for sole purpose of mentioning this show, i just finished watching it. definitely worth a watch but you'll have to go out of your way to find it :/

@kittykittybangbang9367 5 месяцев назад

@@lilithkiwiit sucks that it's not available on any streaming services

@GreatGreenGoo 5 месяцев назад

Its such a shame how little attention Pantheon received. Its such a great show

@blackmagyk 4 месяца назад

just wanted to say the same thing, what a beautifully written show

@kv4648 4 месяца назад

I just finished the 3rd episode and this popped up in my reccomendations

@OhhCrapGuy 6 месяцев назад

It's important to note that even if a brain isn't doing something that can be mapped to computation, computation can simply emulate the atoms the brain is made of. That's obviously more computationally intense, but at a bare minimum, that's enough detail to emulate a brain. It may be possible to perfectly emulate a brain with less detail than that, but that is an upper bound.

@NickV-ez4be 6 месяцев назад

Or just build a brain

@Puerco-Potter 6 месяцев назад

@@NickV-ez4beactually I think your idea will be implemented sooner. If you scan a brain and make a copy with enough precision, you may as well make a new brain for someone

@Builderguy2001 6 месяцев назад

@@Puerco-Potter So then what you would need is: 1. A way to scan the original brain at the level of detail required. 2. Replicate it perfectly terms of neuron positions, connections, and active synapses. 3. A way to interact with (and more importantly keep alive) the replicated brain. TBH, the 3rd step to me feels a lot more simple to accomplish. Might even be very reasonable with the technology we have today. Heck, if people are pursuing digitizing brains as a method to "live forever" or something, step 3 would extend their lifespans anyway, avoiding the complicated debacle until a later point. As some people point out, there are a few ethical dilemmas beyond the scope of the idea you can replicate a human brain. I mean, how does that work with the concept of consciousness? Is it closer to a clone who is their own individual with shared memories, rather than the exact same person who is simultaneously alive and data? But the whole concept of lifespan extension is an ethical dilemma. Living creatures aren't designed to live forever. They're designed to grow up, live, replicate, and die. Cutting death out of the process negates any kind of reason or rationale for any of the other steps. Why replicate if nobody dies? Why grow any more as a person when your capacity as a human is no longer needed? You could slap digitized brains into a satellite and let them live a meaningless existence as data floating around the sun until it explodes or the components degrade. Adding on arbitrary limitations might work of offset some of that, but you're inevitably going to get people who want to bypass that limit. What if someone were to be strapped into that previous satellite and forced to endure some kind of digital torture until the end of time? The worst we have on Earth is a life sentence, but what does that mean when you could casually punish someone forever? Living forever might look like some kind of boon to the ambitious, especially if they want to live forever with a purpose. However, it can also be one of the greatest punishments we could ever conceive. Sorry to go on a bit of a rant there. Medical ethics is one of the most important ethical fields we have left, and I've been finding lately that people are far too willing to toss it aside in the pursuit of some gilded future. Often times, we only see the true ugliness of ideas only after we implement them...

@nodrance 6 месяцев назад

not necessarily. It's possible that subatomic particles play a part, and that even something like virtual particles will have a tiny but noticable effect.

@bennythetiger6052 6 месяцев назад

@nodrance as our knowledge of quantum mechanics increase, I see a lot of neuroscientists actually starting to speculate the role of of elementary particles in the brain. Also, we don't know to what extention the processes are random or controlled, so we might just be utilizing macro-structures as basic processing units, but it might very well be the case, as mentioned in the video, that even atomic structures might be responsible for controlled determined deliberate computation. That would make what we think we know about neuroscience look like kids play

@Raven-ph4uh 6 месяцев назад

“From the moment I understood the weakness of my flesh…” Ngl though if I could make sure that it was truly me in the machine and not a clone, I would be all for this

@royce_beyer 6 месяцев назад

Realistically an emulated version of you would be more your child then your clone

@gabrote42 6 месяцев назад

There's not much difference for the most part. All or most transfers are copies. The question is wether you destroy the original. The copy would not know the difference. FOr a good understanding on how this may work, I recommend the horror game SOMA.

@thorvaldspear 6 месяцев назад

Even if it was just a clone I'd still go through with it

@Raven-ph4uh 6 месяцев назад

@@gabrote42 It’s a very interesting concept, I think another comment had the right idea to slowly phase out neurons for digital copies as a way to transfer a consciousness, but as of now it’s just speculation whether or not it is achievable.

@gabrote42 6 месяцев назад

@@Raven-ph4uh Ship of theseus argument. I'd say it is not any different. Especially with how the brain handles losing half the neurons in itself. SOMA explains it better tho

@AvaRainshadow 6 месяцев назад

The fact that we are getting closer to the science fiction goal of digitizing a human brain is both fascinating and deeply unsettling.

@DobleWhiteAndStabley 6 месяцев назад

"uploaded 1 minute ago" never clicked so fast in my life.

@TheTarged 6 месяцев назад

This topic is explored beautifully in the fantastic show Pantheon which just wrapped up. It is relatively unknown and highly underrated. All the topics in this video and many more are explored in depth through a rich and complex story. Furthermore, It is rare for me to give such high praise to a piece of media, but Pantheon is one of the greatest Sci-Fi shows ever made. I want to get the word out, so people finally discover this hidden gem.

@tintrupeljak3147 6 месяцев назад

The amount of philosophical content that a channel can squeeze out of just the first season of Phanteon is so enormous it almost feels shocking somone didnt do it already

@Paraselene_Tao 6 месяцев назад

I'll check Pantheon out. You remind me of how I sound when I describe The Talos Principle (2014, philosophical puzzle game). There's a Talos Principle 2 game that just released, but I haven't played it yet.

@TheTarged 6 месяцев назад

@@tintrupeljak3147 And yet, no one seems to be covering it. The videos about Pantheon have a few thousand views at most. Which is criminal for such a great show. Yeah, Pantheon combines so many Sci-Fi, futurism, and philosophical ideas while 1). Telling an amazing story 2). Exploring those concepts in depth 3). Having a satisfying conclusion (which many animated shows don’t get nowadays) Unfortunately the big shots at AMC pulled the seasons off any streaming service and it is only accessible now through Amazon prime in Australia and New Zealand. Yar har, there still be ways to watch it though.

@user-fx4dy1zw1b 6 месяцев назад

That cute little puppy character keeps making me watch the videos. They are so adorable

@TanyaLairdCivil 6 месяцев назад

"Time to flip the switch on and say hi to a new kind of human." INCOHERENT UNEARTHLY SCREAMS from PC speakers.

@Vexwisval28 3 месяца назад

Im terrified that if we get this wrong, we’re gonna bring into existence, a sentient, but corrupted being, that actually has feelings and senses, and would be perpetually trapped in some sort of torture chamber. Imagine that but add that it can be copy and pasted, a person would be able to producing mass genocide, on a click of a button

@johnpaulcross424 6 месяцев назад

Glad to have my daily dose of existential dread thanks to man-made horrors beyond my comprehension delivered in such a nice package!

@kx4532 6 месяцев назад

ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-q6wOu02WoIk.htmlsi=Pp-CC5Cg2HYFS-Da

@nicoruppert4207 6 месяцев назад

What is existential dread inducing on this video? Doesn't seem very dreadful to me.

@thesenamesaretaken 6 месяцев назад

@@nicoruppert4207would you like to wake up tomorrow as a brain in a jar?

@VladimirLlyichLenin 6 месяцев назад

This is great, actually.

@jxnrousseau 6 месяцев назад

Would “flipping the switch” end up killing the biological brain?

@joz6683 6 месяцев назад

This channel goes from strength to strength. The topic range is broad and covers topics I would not be interested in. However, they turn out to be some of the most interesting and enjoyable uploads. Thanks to everyone involved for your tireless work.

@smitchered 6 месяцев назад

The rate of uploads is going up! It's also very gratifying to see I video I know will be high quality, but is also under a thousand views because it just came out. Also, I'll take whatever I can get, if supercharging vague neuralink/BMI/mind upload tech is a way to get through alignment, yes please.

@smitchered 6 месяцев назад

Shout-out to Epic Mountain! They make the Kurzgesagt music as well! I especially liked whatever bagpipe sounds were stuck into this one. You should maybe put the Epic Mountain credit above the patron list instead of below? Given these are entirely new songs made from scratch for this video specifically, that would seem natural to me.

@Drengodr 6 месяцев назад

While Epic Mountain did excellent work on this video, Rational Animations almost certainly commissioned them for it, and so should be able to credit how they wish (if credit wasn't specified in the commission contract). If Epic Mountain had collaborated for free, I would agree with you.

@roermy 6 месяцев назад

This is an old music track of theirs, though. Epic Mountain did not make it just for Rational Animation.

@Paraselene_Tao 6 месяцев назад

I was imagining the ethics of this process. We're making a virtual, wholly-functioning brain, and we will be doing this a lot: perhaps billions or more of times. Each time we turn these v-brains on, we're making a potentially conscious, virtual organism. Maybe we can speed up the rate of their life (similar to how we can speed up video games). Maybe it would be too taxing on our resources to let them live forever, so we let them live for a hundred or so years in their virtual yet very realistic universe. As I was writing this, it became clear that I could be talking about myself or ourselves. We could simulate brains in a virtual but very realistic world. Spooky. Anyhow, the ethics of this whole thing is strange because we would be making billions or more of virtual lives that live: suffer, pleasure, boring, exciting, wonderful, and terrible. What's more, we could be testing the v-brain for something particular. Coming back to me or us being simulated v-brains: we could have already served our purpose, or this right now is our purpose, or maybe we haven't reached the point of our purpose yet. Yet another situation: perhaps our purpose is plural or manifold: it could overtime somehow overlap with itself. This whole thing is really bothering me in some existential manner. If I were a v-brain, then I have to say thanks for making me. You (the thing that simulated me) already know that I kind of hate living, but I also enjoy living. Thanks for giving me a life-dream that was okay. It hasn't been too terrible of a life-dream. Blue from the Cowboy Bebop OST is playing in my head, but you already know that, don't you. 😁 I'll see you later, Space Cowboy.

@RorikH 6 месяцев назад

Yeah, AI via simulating an entire brain is a weird but interesting method, and you could make it smarter by speeding up the calculations or simulating a larger brain.

@rickroller1566 6 месяцев назад

They probably don't know that, reading consciousness may (or may not) be harder than making the entire brain.

@nikibasilius4787 6 месяцев назад

My thoughts exactly!!! Now I don't have to write them!

@mooseyard 6 месяцев назад

Look up the short story “Lena” by qntm for a rather horrifying perspective on running multiple copies of one emulated brain.

@krzysiekbudzisz4572 6 месяцев назад

If you would like to be immersed deeper in this idea, there is a game called "Soma" about such poor simulated brain

@hythan9198 6 месяцев назад

I have to say props to the animation team. The narrator's expressions are so lively and cute it creates a wonderful contrast with the serious subject matter. This goes for the recent Genie episode as well. The animation has always been good, but these two episodes have been really standout!

@ncedwards1234 6 месяцев назад

Alternarive approach: 1. Scan (same first step) 2. Represent your brain's history (but not necessarily all at once) 3. Program stem cells to develop in a similar way as you did, potentially including memories For the next decade or so, it'd be impressive even just to guide the stem cells to mimic your personality without major unintended defects, but that would basically be having a kid in a less efficient way. If we get true clones with memories via brain decoding and cellular encoding, you could live forever without a memory lapse any longer than a week (if you always go to weekly checkups). Pro: Immortals thinking more long-term because of centuries of experience and inability to push burdens to future generations Con: If we ALL become immortal, have kids, and continue living on Earth; land disputes will get infinitely worse unless we sever one of the 3 values above

@ArchieGames586 5 месяцев назад

I mean, if we're immortal by that point, may as well move to space, right?

@brucetete3643 6 месяцев назад

If y'all are into the idea of uploaded intelligence I recommend this show called Pantheon, it's animated, has Paul dano, and is super philosophical and cool. My favorite show in the past couple years

@Isaac-fo9rh 6 месяцев назад

The production quality here is on a whole other level. The writing too, just wow

@Mastermind1776 6 месяцев назад

I am a huge fan of your topics, animations, style, and choice of video avatars! Keep up the excellent work!

@gabrote42 6 месяцев назад

I may sorely miss standalone Robert Miles videos, but this is pretty cool. Also, been doing some huge Isaac Arthur binges so this is great

@ate_my_wheaties 6 месяцев назад

I’m so glad I found this channel early on. I love the variety of content, its presentation, and especially the cute doggos. Thanks for the hard work, and keep it up!

@nikhilsourav2025 5 месяцев назад

Recently found out your channel. Your animations and storytelliing are in a league of their own. Happy to be a part of your subscribers keep it up!

@royalcities 6 месяцев назад

Its wild how much computation happens with the brain compared to how little "power" we use. Probably just shows how much work their is to improve - even looking at current ANN's our brains are WAY more energy effecient. meaning there must be some piece to the puzzle we are missing.

@BGBTech 6 месяцев назад

I am left to suspect that what we may "need" may actually be somewhat less than a neuron-level emulation, but rather a way to mimic the general propagation of signals at a moderately high resolution. But, yeah, we may also need something different than current neural nets (which are effectively glorified dot-products with a final non-linear step glued on, such as ReLU or similar). Say, for example, likely, one will also need mechanisms to dynamically adapt neuron weights and to "DC bias" the inputs (say, having the neuron adapt each "synapse", say, so that the average DC-biased input is around 0, and the weights such that the normalized inverse-weight range is around -0.67 to +0.67; independently of the weight which goes towards the output). Then one can have a sort of running "inhibitory state" that flows backwards through the net as a sort of mimic of the effects of back-propagation, along with a "running average" of its current activation state (and incrementally adjusting the weights as it goes; so the inhibitory state flows more strongly towards neurons with a higher activation state; with inhibitory signals also being generated as neurons go outside a range of, say, -1.0 to +1.0). In some small-scale experiments in my case, this sort of thing seems promising, but is more computationally expensive than the more traditional strategies (and doesn't depend as much on an externally imposed "training algorithm"). Well, and admittedly I have tended more towards a "signed square root" transfer function (rather than ReLU or similar). There may be some other parts needed as well, I don't know. But, in any case, I am still pretty far from anything "actually useful" here... Note, have mostly done experiments with custom code in C, not using any mainstream NN libraries (mostly in the form of small one-off experiments).

@jacobmagana4258 6 месяцев назад

its because 1. brains only have to perform computation when a neuron spikes/fires, while anns process the entire brain at once. additionally, brains have short term memory built in, while anns are one-directional and so only have access to long term memory from training each time theyre run (because of this they will perform a lot of waste-full calculations, if theyre trying to solve a problem that requires thinking ahead they will have to re-come up with the same plan over and over). SNNs solve both these problems because they actually mimic the way that brains work on a neuron to neuron level, and as a result are dramatically more efficient than ANNs (though theyre less predictable and harder to train) 2. brains are purpose built, the software has been abstracted to the point where it basically is the hardware, rather than being emulated on something else

@jozroz2165 6 месяцев назад

The animations have gotten so fluid and detailed, incredible work!

@FroggytronFrog 5 месяцев назад

These videos are so strategically excellent and underrated! Also given the animation and commentary, they are easy to understand. Great job here Rational Animations

@raylo555 6 месяцев назад

If you want to digitize a human brain "Ethically", you have to do it through what I've pet-named the "Theseus Method". Using the Ship of Theseus Paradox: Gradual, segmented replacement of the human brain will maintain the illusion of a single consciousness, instead of creating a duplicate and destroying the original. This will require a communication between the brain and the device you're being uploaded to, along with the ability to; remove, replace and emulate the parts of the brain that are gradually being removed but this also removes the moral dilemma of "I'm just making a copy" since the illusion of continuum is unbroken.

@williamervin3272 6 месяцев назад

Exactly! Such technology is way beyond brain scanning, but maybe with life extension and ASI running the show, we could live to see the day where such a thing is possible.

@Dreadon1 6 месяцев назад

Look up old man's war. Its a scifi book that used this exact method to upload people's minds. Fun book too.

@ilrisotter 6 месяцев назад

This is probably the only ethical way to do this. Otherwise you're just imposing a lifetime of memories on a completely new individual.

@jacobmagana4258 6 месяцев назад

but why? like seriously what is the difference unless you believe in a soul/stream of consciousness (which both require something metaphysical)?

@jacobmagana4258 6 месяцев назад

@@ilrisotter if it has your exact brain structure then how is it a new individual (unless, again, you believe- without evidence- in a soul)?

@cloakedoblivion22 6 месяцев назад

I feel like with time this channel is becoming more and more unhinged in like a psychedelic sense that I like ❤️

@neptunite5973 3 месяца назад

we all love the animation, and how well the concept is explained. But i also wanted to point out the lovely sound design! Love all the little beeps and boops

@armordcrab 6 месяцев назад

Thank you!!! Keep trying to explain this to people but it's hard to lay it all down in casual conversations ❤

@cptn.penguin902 6 месяцев назад

Interesting content as always! But I would also like to point out, that the animations keep getting better and better each time. Kudos to the animator/s!

@OhhCrapGuy 6 месяцев назад

"Why is the simulated human screaming?" "Did anyone give it sensory input?" "... oops"

@CalvinNoire 5 месяцев назад

"I CAN'T BREATHE, I CAN'T TOUCH, I CAN'T SMELL, WHAT THE F**K IS GOING ON, SOMEONE HELP ME" "God damn it Jarrett! I knew we should have given her a virtual environment to be in!"

@RandomVex 6 месяцев назад

Wow, that sounds truly fascinating I sure wonder how the elite class will use it to gain more power over everyone else

@lemonjuce13 5 месяцев назад

I just came here because I rewatched your epic conway's game of life video, as I'm currently tasked with implementing conway' s game of life in form of a C command line tool. I saw your updated Description about how you started creating content gain and I have to say youre doing really good! Kurzgesagt-style videos about Tech topics? It doesnt get any better than this! Earned my sub in half a minute

@thebirdbrained. 6 месяцев назад

This was so incredibly well made. Deserves millions of views and at least that many likes!

@muwgrad1987 6 месяцев назад

Bravo, team! Great work!

@razzledazzlecheeseontoast9808 5 месяцев назад

Watching Pantheon S2 and I'm really glad this video was recommended, feels good to have some real world info to contextualise the sci-fi.

@acidcosti19 6 месяцев назад

I love this channel and every backround is a masterpiece

@RodBlanc 6 месяцев назад

I was watching Pantheon season 2 yesterday and this video came in the perfect time for me

@N.i.c.k.H 6 месяцев назад

The key thing that I take from this, that people seem to have missed, is that we are likely to have low resolution uploads long before we can do anything that might be called a "true", lossless, upload. This obviously raises huge ethical issues.

@JustWojtek 6 месяцев назад

Great video! Love the animation style of all your videos and the topics you tackle. Although I don't doubt that humanity will be able to upload a brain at some point, I doubt that it will coincide with uploading a mind. We use the terms interchangeably, but crowning the brain the seat of the mind has always been more of a (reasonable) best guess then an actual fact. I recommend reading Anil Seth's "Being You" on that matter.

@user-on6rv4ik6q 6 месяцев назад

You guys just keep adapting my favorite papers

@Volvith 6 месяцев назад

I can't wait for the inevitable _"You wouldn't download someone's mother"_ anti-Piracy PSA's. 😂

@Pheonix1328 6 месяцев назад

The third step is by far the easiest. At the most basic level, it's just a matter of scale. Even if it means slowing down, since it takes longer for one side to communicate with the other (unless the architecture changes, ex. light based computing), you can always build a bigger computer with things like a Matryoshka Brain being an extreme but realistic example. Step two is harder but depends a lot on the information gathered in step one, which I think will be the hardest step. All I know for sure is, you couldn't pay me enough to be the first person to get their brain scanned. "Well, the scan is complete, the data compiled, and we're ready to start the simulation. Do you want to talk to your digital self?" "Sure," they nod looking nervous but excited. The simulation starts and horrific screams blare out of the speakers. Everyone is frozen in fear. The screams continue for a few seconds, then silence.

@dangerfly 6 месяцев назад

A seamless transfer of consciousness require neurons to be gradually replaced by artificial ones to maintain continuity otherwise we run into the Star Trek Transporter Problem.

@Kubovec20 6 месяцев назад

Yeah

@kx4532 6 месяцев назад

Why, the you that you were 10 mS ago is dead and the you that you are in 10mS will be a completely different you even is it gets cranking on a different computer. The transporter problem is siperstition.

@smitchered 6 месяцев назад

Only relevant philosophically, I think, which isn't the approach of this video. Ship of Theseus is what you mean? Is that essentially the same thing as the Star Trek Transporter Problem?

@kennyholmes5196 6 месяцев назад

@@smitchered Yes and no, as near as I can tell. Trek Transport is pretty much the same as Ship of Theseus, but done all at once instead of over time, and using the original boards to measure out copies with.

@kx4532 6 месяцев назад

@@smitchered It's not the ship of Theseus problem, it's the Bill's laptop problem. Replace the screen, same computer. Replace the keyboard, same computer who cares if the keys are red now. new ram chips, works the same, same computer. Replace the hard drive, whoops. Image the hard disk to the new drive, same computer again. Image the disk to a virtual machine, works the same, Same computer. It's the Church Turing thesis.

@leox8038 6 месяцев назад

That's the reason why I started studying bioengineering, now that I'm close to get the degree I seek other interests, but the fascination with this topic still remains.

@charlie8458 6 месяцев назад

Training an AI to "copy" the personality of someone seems more achievable than mapping a brain and running the code you get out of it. It seems such a dangerous and weird technology though. The show "upload" is a good example of this and yet, it doesnt explore the terrible possibility of having a dictator copying himself and just living forever this way. Right now, life allows us to hope that one day that dictator will be dead and things might change. What if they never die ?

@longline 6 месяцев назад

The nuance and detail you all bring to this is excellent. Black belt, top notch. You covered my every caveat, and I have a lot of them on this subject. You all are masters of both depth and concision.

@DeGameBox_SRBT 5 месяцев назад

the worst thing about this is that at the moment we can't prove whether we will be in the simulation ourselves, or only our clone.

@milkibearmilkibear 6 месяцев назад

Another amazing video, keep up the great work!!! It would be very nice if you make a video on the Dark Forest theory, or the Killing Star with it's relativistic bombs!

@ryanedgerton1982 6 месяцев назад

There's one additional problem: emulating "peripherals". Any emulation of another computer system requires simulating hardware-to-hardware interactions that aren't there but were in the original -- think the Link Cable for the Gameboy, or the joysticks on an Xbox controller. This is comparatively easy when emulating one manufactured system on another, since it's possible to map peripheral connections onto similar hardware, but we will have no such easy parallels for the first fully simulated brain. It would be akin to sensory deprivation, which would likely cause the emulation to break down (since sensory deprivation drives people insane over time). While we might be able to address hearing and seeing with cameras and microphones, the sense of touch and the need for movement is a very different issue. Even if we upload these simulated brains into robotic bodies, we're going to have to contend with a host of psychological phenomenon that could foul the effort. Hunger, for example, is chiefly a biological imperative but also has very strong psychological roots. What do you do with a emulated brain that's developed a sudden craving for tacos? Or one that risks getting neurotic after not feeling the soothing cleanse of a hot shower for several months? If we create an emulated brain without memories, how do we address the infant mind's need for physical comfort (without which they can, in fact, die)? To accurately emulate the human brain on a computer, we will also have to figure out how to emulate the human body and a human-compatible environment, and at that point we've reached full-immersion VR akin to The Matrix. The original goal, and whatever we hoped to actually do with it, will be completely eclipsed by the secondary efforts needed to make it viable.

@devinward461 6 месяцев назад

I agree that we'd need to simulate a virtual environment to make mind uploading worthwhile, but I don't know if it'd be harder than achieving uploading in the first place just by virtue of how utterly monumental a task it'd be to upload a human brain to begin with

@andreyrumming6842 5 месяцев назад

By using Parallel Random Access Memory, the issue you mention in 8:23 about processor/memory bandwidth is already solved, with a catch. The issue with modern processors is not just processing speed, but data throughput. If your processor is taking half it's time just to move around memory, it's gonna be bottlenecked pretty badly. Especially if you have hundreds of processing cores trying to access memory at the same time, having to wait for the other cores to finish their memory operations before you can do yours (A critical issue that stops supercomputers from just piling on more processors without having segmented disk memory). PRAM however allows you to have one memory bank, and just pile on the processors arbitrarily. But here's the catch, the more memory you need, the more physical pins you need. For 1MB of PRAM, you need around 1.2 million pins, or a grid of 1,095 x 1,096 pins. For a system like the human brain, we'd likely need petabytes of memory, which in PRAM would require a grid of 32 million x 32 million pins. It'd work, sure, but it'd get big QUICK. It does also mean that if you had 1 neuron for 1 core, that'd require something close to 86 billion cores. Not that bad now that our PRAM allows us to slap on processors without any extra complexity, but still. And each core would need it's own separate memory, mainly for storing it's connections, although this could be standard cache memory nothing else there

@michaelmcwhirter 5 месяцев назад

These videos are always great. 💪

@nikibasilius4787 6 месяцев назад

I just wanted to say i love your channel❤

@hasamat38 5 месяцев назад

Found you channel today, Good video!

@WhatWouldVillainsDo 4 месяца назад

Just binged Pantheon and had to look intontge science of it, live the video!

@PrinceNewRomeo 5 месяцев назад

I don't know how you two feel about this, but I'm praying for y'all. Loss is just...so very hard. I'm sorry for your loss.

@leo.skilled 6 месяцев назад

love your videos hope you'll get more subscriber becouse you are the person that definetly deserves it

@anthonylipke7754 6 месяцев назад

I suspect at least some kind of distributed architecture memory and compute near one another if not combined will be prefered. AI chips often have large cache and high performance memory on package. The architecture would likely be different than traditional computing. I expect imitation to be easier than full emulation. Simplified to run for high level not low level similarity. Directly interfacing neurons and computers offers some interesting options as well.

@ludicrouS_406 6 месяцев назад

running the emulated brain as a program might result in some issues regarding the senses. since the brain has neurons that work directly with sensory organs such as the ears, the nose, the eyes, the rest of the nervous system, etc, it would be necessary to provide the emulated brain those organs (maybe a simulated version of them), or manage to find a way to cut those neurons off from the model of the brain, which would most definitely lead to complications (not to mention that by doing so you'd be cutting any connections to the outside world for the brain, therefore not being able to really tell if the emulation even works properly in the first place)

@somerandomgal3915 6 месяцев назад

Considering that a few parts of the brain have the specific job of regulating the brain chemistry and relasing dopamine and seretonin accordingly, but those themselves also being in need of some other biochemicals for that... and then *also* figuring out how to implement/simulate those within our emulated brain programm on the computer... While it technically isn't necessary to know how every complex bit and detail of the brain works and operates, knowing more of that *would* help making the data collecting bit as well as the programm writing bit more efficent/less ressources intensive and by extension also the running the emulated brain part. That emulated brain itself would most likely end up running on fuzzy logic (which most neuronal networks and machine learning programs work with), which would need to be applied for each and every emulated neuron within that brain. That being said, I'd be surprised if we ended up running that emulated brain on a computer less complex than it itself. And that all isn't even considering the impacts of very limited/controlled/selected input data (sensory or otherwise) on a full emulated brain, cause i for sure wouldn't like the idea of kinda being within a void of not seeing, hearing, feeling, tasting or even smelling while the emulator of me runs, expect for the few inputs I'd end up receiving from the person running me. Living brains tend to hallucinate in the extended absence of sensory input after all.

@deleted_handle 6 месяцев назад

Thanks really needed this tutorial

@fritzbender5916 6 месяцев назад

Really high production value and interesting topic

@spoke2639 5 месяцев назад

One potential problem I see is that a big portion of our brains run and regulate thousands of automatic functions in our body that we don’t really have any conscious awareness of. A computer wouldn’t be designed to mechanically perform any of these bodily functions, so as soon as we turn on the program the Artificial Consciousness will start freaking out because it isn’t receiving any signals that the bodily functions are carrying out their instructions.

@user-me8sj2sw4r 6 месяцев назад

The only bad thing I can think of is: We have no way to transfer the mind to a computer, rather we only know how to create a mimic of said mind. Though still neat… not exactly helpful if you the individual want to stay alive a lot longer… I wonder if a pathway to that idea can be made. At least a theory perhaps. -.-

@krabman6297 6 месяцев назад

I personally think we should consider what I'd be like to exist as a brain simulation. Without constant information about senses, bodily position, and status, how would a brain react. It's almost as if being in a sensory deprivation tank but on a more absolute level, we should consider the emotional and psychological effects of running a brain for long periods of time in this state or the ethics of even having a digital copy of someone that can easily be duplicated, destroyed, altered without consent, or subjected to experimentation.

@Giganfan2k1 6 месяцев назад

My body is trying to kill me... Of course I want to upload my mind. XD

@AlkisGD 5 месяцев назад

I was hoping the video would go into more detail regarding sensors (emulated or otherwise) and other parts of the body that affect the simbrain. For example, imagine putting the simbrain in a virtual reality Indian forest. How's the simbrain going to react to an approaching tiger if we haven't accounted for the fight or flight reflex? And how are we going to implement the reflex? Simulated adrenaline, etc.? So many things influence the brain via substances (hormones, neurotransmitters, proteins, etc.) that would also have to be accounted for and possibly simulated. Our individual gut biome, for example. I also imagine there are countless feedback loops we're unaware of because our consciousness doesn't have access to them. Someone says something, stuff happens in various parts of your body, and boom, you're starting to feel anger rising within you, your heart rate goes up, your palms get sweaty, and of course your thoughts are affected as well-it's how crimes of passion happen after all. That's a lot of stuff to simulate for the simbrain to behave like the person we're trying to digitize.

@time.window 6 месяцев назад

What if we used nano bots to substitute the organic material in our brains with something synthetic analogue to it? Essentially slowly turning the brain into a computer

@SumiEwiets-idgaf 6 месяцев назад

I'm currently pursuing CSE with Specialization in AI and Machine learning while also doing another Bsc bachelor in Data Science.I'm still a teen so I've much to learn but Hopefully I'll be part of such revolutionary projects.

@HalloGaming 6 месяцев назад

One more thing to note about "Computer Power" is that Moore's law is closing in on the "we can't make transistors any smaller," however quantum computing might be an even more efficient method once that technology evolves. Although I will note that quantum computing as a successor to digital computer is very theoretical.

@Rawi888 6 месяцев назад

I really love this channel so much. I'm going to keep recommending it to everyone as long as you guys keep contributing to better humanity. ❤️

@tukkajumala 3 месяца назад

One major hurdle is that we do not know how much of thinking happens inside our brains and how much of it works like an antenna. Brains may receive external signals in ways we can not yet even detect, let alone measure or simulate. There is so much our current science considers "impossible", which we should still somehow take into account, before we could hope to simulate a full human experience. Before we dare to approach the impossible, we will remain stuck.

@and_I_am_Life_the_fixer_of_all 4 месяца назад

Feels like this video was made for me! Can't wait to get it working by 2055! @Rational Animations please do one about aging also, the continues experience was not addressed in this but it was interesting!

@oscarbrontide7119 6 месяцев назад

I feel like leaving out consciousness is a bit of a mistake? Why on earth would you want to scan your brain? For digital/robotic clones? I’m much more interested in how we would go about this if we wanted to keep our continuity/consciousness intact. How do we upload a consciousness without breaking it?

@rowbot5555 6 месяцев назад

slow brain modifactions where we slowly replace parts of the brain with synthetics over the course of multiple surgeries

@w0tch 6 месяцев назад

We have a lot of important neurons in the spinal cord, in the belly, and at other places in the body. To emulate the full human experience, you need to emulate the body as well. You won’t get the sense of scent or taste, but the other senses may be doable. It is important that the brain emulation is well connected to it’s environment for it to function, or you need a very rich environment simulation as well. But my biggest question is: if we can upload our mind, we keeping it brain like ? If we can become a machine with endless possibilities, why stay human ? 😄

@GreatGreenGoo 5 месяцев назад

Pantheon and the Bobiverse books brought me here. UI is very intriguing. I do wonder if quantum effects have influence over neurons and other such systems. If so emulation just got infinitely more complicated.

@nemonomen3340 6 месяцев назад

It’s also very important to figure out how to stimulate the whole brain emulation (WBE) in such a way that the individual doesn’t go insane. The exact requirements are unclear, but I imagine a simulated body that feels like it’s breathing is absolutely required. Sight and hearing would be very important as well and at least one should be included from the beginning.

@tommysalami420 5 месяцев назад

o.o

@liaml3291 4 месяца назад

If you want to dive into the ethical implications of this kind technology through a fictional lens, I can’t recommend the show “Pantheon” enough. It’s gotta be one of my favorite pieces of media to release in the last decade.

@daniel4647 5 месяцев назад

A lot of our brain is doing "busy work" though, basically keeping us alive, so a lot of it could probably be cut out in order to optimize it for a digital existences where it wouldn't have to manage a body. We'd probably like to keep things like our senses in order to properly simulate them in a virtual environment, but we don't really need a heart pumping or digestive system or any of that stuff. I don't know, but big animals have big brains, and and intelligence seems to correlate to the size of the brain in relation to body mass. We should probably upload a complete one first just to see if it can even be conscious, but once that's done we can start cutting out all the parts which it doesn't need and then we'd be able to learn how to optimize the whole process before uploading, or even scanning. In digital form we can just remove things and restore from backup, so we'd have lot of freedom to just pull things out of a living brain to see what's fine to throw away and what we need.

@Leadvest 5 месяцев назад

The best approach would be to focus on digital twinning, with something complex enough to have testable behaviors, and memories. It's likely that with enough characterization, advanced temporal interference measurement methods will be accurate enough for most purposes. There would also be a lot of abstraction in the emulation, you don't need to calculate derivatives directly in modelled synapse phase interactions, if you can impose results from on-chip analog hardware accelerators.

@loszhor 3 месяца назад

6:36 That's super cool that we've caught up to a human brain!

@GeoffRogers42 6 месяцев назад

Say hello to a new kind of human, who goes insane shortly after due to not correctly emulating the sensory system.

@theallmemeingeye5927 5 месяцев назад

Hugely fascinating video, though I have to admit I am a little disappointed you didn't even mention the existential- and astronomical-suffering-risks that could arise from this technology if we aren't careful. Although fictional, the Black Mirror episode White Christmas and the online short story Lena MMAcevedo do a pretty good job of conveying how horrifyingly wrong things could go. Maybe an idea for a future video?

@Eldagusto 6 месяцев назад

It’s funny I know Anders Sandberg because he used to run a homebrew fan page for the game Mage the Ascension on the 90’s. He’s high quality in his fan projects for his RPG hobby but he is also a world class genius in the professional field in the sciences.

@thismakesnosense 4 месяца назад

Great video👍

@Narutendo3 6 месяцев назад

Great video. And it caused an idea to occur to me that emulating the behavior of neurons and synapses fundamentally doesn’t even make sense without also emulating the entire body, its microbiome, and a vast and detailed enough environment that doesn’t break down by something as simple as walking 100 feet. Basically you need to simulate all of reality to simulate a mind, because we are more deeply interconnected than just a brain isolated in a void. In the limit of a worm, this may actually be possible since its relevant environment is probably very simple. For a full healthy and happy human, I doubt it’s even possible. Consciousness seems to me like something profound that is futile to achieve in a machine, and maybe we should just study and appreciate our biological consciousness more for all that it may be.

@hesperosshamshael2873 4 месяца назад

Some years back an idea occurred to me, and I do wonder if it has some merits for an upload. The premise is developing a Man-Machine-Interface and permanently linking the brain to external/additional processing mechanisms to emulate brain functions and ultimately act as 'expansions', taking advantage of neural plasticity to try and coax the brain into integrating this external thinking machine hardware. Given time, I have to wonder if this could ultimately allow the brain to render its own flesh components redundant and before you know it, the brain isn't even in the body anymore, nor is it organic. I've seen stories and studies showing the brain can do some remarkable adaptation, readjusting to some remarkable damage and loss. So the root of the question is can you get the brain to accept an 'upgrade'?

@Ky-Nas 3 месяца назад

This makes me curious as to how difficult it would be to create a robot body that works as seamlessly with the brain as an organic one does. I would assume that making a body that responds in the same way even down to the smallest scale would be difficult, but I'm curious how much of that work is cut down by the developments achieved during the process of uploading a mind.

@theredsaurian 3 месяца назад

Assuming we're able to accomplish all of that, at best the resulting emulation is something that thinks it's the original person (original person may be still around or dead if the uploading is fatal) but is still only a copy and arguably a philosophical zombie (and will go mad with zero sensory input if we never bothered to program any for it), and at worst it is a broken insane thing that screams in all possible frequencies, or the output of neuron interaction is completely uninterpretable, or activity ceases. I personally don't think brain emulations can ever have continuity of consciousness with the original person or even be conscious in the first place, just very good at mimicking it, if that.

@EnbyNomad 6 месяцев назад

Pantheon just wrapped up so I'm primed for this

@-eternal 5 месяцев назад

I was thinking more along the lines of slowly converting to a mechanical brain so I can still be I instead of having a clone of me lol

@allibababoo 5 месяцев назад

The book "walkaway" does a fantastic job or showing what happens next (maybe, it IS a sci Fi novel after all

@somelaser5906 6 месяцев назад

This is a very interesting video! Another aspect on this subject I'd like to see is more behind the ethics of creating a simulated human brain; Are they actually human? If so do they get all the same rights as a biological human? Can we just turn them off? Is it morally acceptable to even attempt this in the first place? I think this would make a fantastic video that would fit right in on your channel!

@OpreanMircea 6 месяцев назад

Excellent video

@kaielvin 5 месяцев назад

Pretty sure we'll be uploaded before this is possible. It is like thinking that to simulate a wave, we need to emulate all the water molecules (or that to fly, we need flapping wings, for that matter).