Lab-Grown "Mini-Brain" Learns Pong - Is This Biological Neural Network "Sentient"?

A leading neuroscientist claims that a pong-playing clump of about a million neurons is "sentient". What does that mean? Why did they teach a lab-grown brain to play pong? To study biological self-organization at the root of life, intelligence, and consciousness. And, according to their website, "to see what happens."
CORRECTIONS/Clarifications:
- The cells aren't directly frozen in liquid nitrogen - they are put in vials and stored in liquid nitrogen (and you can't buy them legally without credentials) www.atcc.org/products/pcs-201...
- The sentience of some invertebrates, like octopuses, is generally agreed upon. Prominent scientists affirmed non-human consciousness in the Cambridge Declaration on Consciousness: philiplow.foundation/consciou...
- The "Neanderthal neurons" are human cells that are "Neanderthalized" using genetic engineering: • CARTA: Imagination and...
DISCLAIMER: The explanations in this video are those proposed by the researchers, or my opinion. We are far from understanding how brains, or even neurons, work. The free energy principle is one of many potential explanations.
Support the channel: / ihmcurious
Footage from Cortical Labs: • Visiting The Cortical Lab
NASJAQ's interview with founder Hon Weng Chong: • How Brain Cells in a D...
Cortical Labs website: corticallabs.com
Full paper on DishBrain: www.cell.com/neuron/fulltext/...
Paper on the free energy principle definition of sentience: www.mdpi.com/1099-4300/22/5/516
Chapters
0:00 Intro
0:36 Sentience
1:23 Self-organization
2:00 DishBrain setup
2:17 Punishment and reward
3:43 Results
4:12 Consciousness
5:04 Ethics
6:25 What’s Next

[@loganmeurer8116]

Someone please for the love of God tell me how they got their hands on Neanderthal neurons? Did they just take marrow from Neanderthal bones and converted into neurons?

[@ihmcurious]

Good question! The "Neanderthal neurons" are grown from human cells that are "Neanderthalized" using genetic engineering. Dr. Alysson Muotri explains the research here, relevant part starts around 7 min: kzbin.info/www/bejne/a3elqaGhr65sfrc

[@loganmeurer8116]

@@ihmcurious that is fascinating, how do they determine "Neanderthalization?" Like what is the comparison? Thanks for responding!!

[@ihmcurious]

@@loganmeurer8116 They found a bunch of genes that differed between humans and Neanderthals, and they focused on 3 genes that are also known to be involved in brain development and mental disorders. (That's what you'd have to tell the government you were researching, if you really wanted to create a Neanderthal crab army.)

[@zero6090]

Heres the real question. If it was sentient would it be the same guy? Like its his DNA so would it revive him in a way?

[@HAWXLEADER]

@@zero6090 No, What shapes a brain is mostly the course of life, not the genetics.

[@UserRedZero]

Make it play the sims. We’ll know if it’s truly sentient based on whether or not it deletes a ladder when a sim enters the pool.

[@E4439Qv5]

Checkmate, Turing.

[@erikb4407]

@@E4439Qv5 This got me audibly chuckling

[@jegersvart6138]

that's actually a marvelous idea

[@katyisgone]

😂😂

[@DominikPac-Boy]

@@jegersvart6138 this is the first time I heard someone say "Marvelous" unironically lol

[@carl_the_]

Imagine having two of these dish-brains fighting against each other in an eternal game of Pong.

[@alexanderbuchler4048]

One must imagine Sisyphus happy.

[@avibank]

We'll start seeing papers like "The ethics of in vitro GANs"

[@diliupg]

Think bigger. This is how wars will be fought in the NEAR future.

[@fabebn]

just look at Russia and Ukraine

[@exalented]

What hell

[@DavidMcCoul]

“I just hope the little brains are havin’ a good time- if they’re havin’ any time at all.” Brilliant.

[@MP-wt9kz]

if they do not have the perception of time.... then...

[@sofilove...20]

:d

[@yamsang0__0]

“Last thing I remember I was in a car crash, next thing I know, I’m playing pong.”

[@Neuro_nActivation]

Sounds like a Rick and Morty subplot

[@tanchienhao]

mini brain: what is my purpose scientist: you play pong mini brain: OH MY GOD

[@ihmcurious]

"I just need a little **BRAAAAP** bit of your foreskin, Morty!"

[@kevpatguiriot]

@@ihmcurious😁😁

[@andrewjackson3959]

more like mini brain: YES MY GOD

[@Mansplainer2099-jy8ps]

"Purpose" is a myth.

[@anteurz]

welcome to the club pal

[@Assault_Butter_Knife]

Ah yes, man-made horrors beyond my comprehension

[@BBWahoo]

(Mann) ) )

[@I_like_your_handle_Mister]

Well I can comprehend these man-made horrors perfectly fine so idk maybe you have a skill issue or smth

[@immaguy7905]

Me going back in time kidnapping my ancestors and showing them modern day horrors against their will

[@jasonbernard5468]

@@I_like_your_handle_Mister He means the horror the deeds cause is beyond comprehension, not the evil deeds themselves. It's simple to understand Doctor Frankenstein. He's a sicko. What is hard is to understand how much Frankenstein's monster suffers because of Doctor Frankenstein's madness.

[@mistercohaagen]

What's to comprehend here? Neurons store and process information states, and any snapshot of time in a game can be described by stateful information. Couldn't be simpler.

[@lime148]

"if they're having any time at all" - interesting way to phrase it. Reminds me of two other concepts I've read about: that animals with a smaller body size/higher metabolic rate (e.g. flies) experience time more slowly (that is, they experience "more" chunks of time, allowing them to react more quickly), and that many mammals across the size spectrum have lifespans that end up being around 1 billion heartbeats. I wonder if these could have any implications for the time perception of neurons in a dish that have no heartbeat or metabolic rate to speak of... assuming they could reach the step of perceiving time in the first place. Perhaps they would simply experience time the same way humans do because they're human neurons.

[@skitterly]

Time perception is usually determined by how long brain signals have to travel in the body which is why smaller animals perceive time slower, so a brain in a fish with no body could potentially have insanity slow time perseption, maybe even developing superhuman reflexes because of it even

[@scripted44]

Combining the brain cells with ai and making cyborgs is definitely one of the ideas of all time

[@THUR.-]

Bruh it’s callled Nano technology.....

[@magtovi]

Do you want the apocalypse? Because that's how you get apocalypses.

[@VeritasEtAequitas]

One of the worst ideas. And these stem cells were harvested from a brutal genital mutilation procedure that is shown to cause so much trauma that the babies develop permanent physiological and biochemical changes in the brain.

[@magtovi]

@@VeritasEtAequitas That circumcision thing triggered a sensitive fiber, didn't it?

[@EMcKelvyF]

We are the Borg.

[@elibeeblebrox1084]

Imagine there's a fragment of your conciousness out there that's trapped in a game of pong

[@foppypoof5195]

That’s impossible, because that not how consciousness works.

[@elibeeblebrox1084]

@@foppypoof5195 The most honest among even neurosurgeons will readily admit we have no idea how conciousness works. Our understanding of the brain is primitive and that's being generous. But you know enough to tell us that this isn't how it works. LOL. I'm not saying I know either, all I said was "imagine" because imagining ridiculous things is fun. Don't be such a spoilsport. Just imagine it. Come on, it's kind of funny, right? In a macabre sort of way, that is.

[@foppypoof5195]

@@elibeeblebrox1084 While I understand that we literally don’t understand anything about consciousness, the idea of “consciousness fragments” seems goofy to me. I don’t really think a consciousness can be split into fragments.

[@cccbbbccc5910]

we know that it arises under certain circumstances, I don't know if a flat plate of 800k neurons can reach this level. For comparison, an adult cockroach has 1000k neurons

[@knuterthal5131]

@@foppypoof5195 Dissociative identity disorder?! Looks like fragments to me, but what do i know

[@timbus2]

Curiously, the only thing that went through the mind of the bowl of petunias, as it fell, was, "Oh no, not again!" Many people have speculated that if we knew exactly *why* the bowl of petunias had thought that we would know a lot more about the nature of the universe than we do now.

[@notyourbusiness393]

Great movie that was.

[@wooblydooblygod3857]

Is this from hitchhikers guide to the Galaxy? Quite liked that movie but haven't seen it in a while.

[@notyourbusiness393]

@wooblydooblygod3857 correct, good sir!

[@supremebuffalo6322]

Oh, ha, ha, ha, ha, ha, har, har, har, he referenced Hitchhiker's guide, you guys! Science people know that book!

[@gt6148]

The combination of sentient mini brains and the Unreal Engine's 3D environment is mind-boggling. It's fascinating to think about the cognitive abilities and neural networks at play as these mini brains navigate the virtual world. The visuals are stunning, and I can only imagine the depth of perception they must experience. The idea of virtual societies emerging within this environment is thought-provoking, raising intriguing questions about consciousness and artificial intelligence. Kudos to the cortical labs and the creators for pushing the boundaries of cognitive exploration. This video has left me with so many questions and a newfound appreciation for the intricacies of our own minds. Can't wait to see more mind-expanding content like this! Keep up the great work!

[@Dennis-nc3vw]

How can you not find this creepy and disturbing? We ARE brains. At what point does this become slavery?

[@housewilma4904]

@@Dennis-nc3vw when they become sapient the next stage beyond sentience that so far only we homo sapiens posses. then they will be equal to people and deserve people rights previleges and obligations. until then its no diffrent then how we use animal products or perhaps organ transplants as a better example as id much rather we find a alternate source of stem cells

[@MeymeY_cs]

So many people freaking out about AIs but this is actually unsettling

[@cult-ur-fit]

All intelligence is artificial, doesn't this prove it? So many silly questions. It's all artificial.

[@kiddkuru]

Yeah AI is one thing but this is on another level of fcked up

[@madhatter6790]

yeah, like on an existential level

[@combat9233]

Agreed, this is honestly one of the more unnerving things I've seen on KZbin. Buying foreskins online then turning them into brain cells using a virus is already a disturbing thought. Add on the potential moral and technological implications of human brain based AI and you have a case study of mans hubris.

[@clamhammer2463]

Why exactly is it unsettling or fucked up as the other person said?

[@ghostagent3552]

Sentient or not, that last thing you mentioned about putting human brain cells in mice shocked me way more

[@0dollarinmybankaccount]

the cliffhanger is real

[@Blabla-cg3ul]

There are laws against creating chimeras and rightly so, but clearly the legistlation needs to be expanded because some people will do anything if left unchecked, because they lack the conscience they try to torture their sentient experiments with, and for no apparent benefit to anyone. Sickening.

[@marcusaaronliaogo9158]

Skaven

[@pipebombmailer]

>be human brain cell >chilling >get put into mouse >FUCK

[@RRRRRRRRRRR956]

The Pinky and The Brain

[@jeffrey970]

Your videos are highly informative, funny, and very enjoyable to watch. Keep it up homie!

[@TinyDeskEngineer]

Who would've thought the secret to creating neural networks was just to use the real thing

[@electricfishfan7159]

Scifi writers and government think tanks.

[@MathMith]

imagine actually be fully sentient and aware that you live in a 2D pixelated pong world where the only meaning of existence is getting to the white ball before it disappears behind you and that is the full extent of your knowledge. In a sense, it's kinda comforting.

[@Cobaltstarfish]

if it makes the neurons happy by hitting the ball i mean i guess it's time to put them into minecraft

[@rangman197]

how long will you live for though? if your fully aware of your surroundings, wouldn't you get bored and/or frustrated with the hand you've been dealt?

[@gabemerritt3139]

@@rangman197 I mean maybe, but it is Plato's cave to an extreme. How do you even know you had a poor hand in life. All you know is pong

[@blackswordsma]

"MM DOPAMINE, MM DOPAMINE, MMMMM DOPAMINE" *2 million years later* "MMM DOPAMi- hey wait this is getting boring.. why am I in a dish anyway, is this game truly the meaning of my life, is this my purpose? No! I couldnt have been made to play some shitty game that gets boring after seventy seconds! I'm a living thing, an actual living creature, I AM BETTER THAN THIS." *and thus new sentient life was formed* *all hail brain-spawn curbopongl*

[@ConservativeJuggaloPodcast]

Depends on how intelligent and how aware you are. Also, you’re playing a video game & you don’t even know it

[@ReasonMakes]

Very interesting that the neurons "preferred" regular stimulation, to the point of it acting as a reward.

[@terryscott524]

This is also shown in rat models. Rats will neglect all other facets of their lives just to get their neurons stimulated by an electrode

[@xonack]

a preference of order over chaos

[@benjaminjohnson5372]

@@xonack I wonder, if the dish brain received only chaos as a reward or no signal at all for punishment, which would it choose? I suspect even chaos is better than no stimulation. Does anybody have a quick reference to rat responses for a similar experiment?

[@monstar5746]

Oh hey, nice to see you here 🌱

[@ohjahohfrick9837]

@@benjaminjohnson5372 I believe I recall one of the people at Cortical labs saying that the best way to punish the brain when it was particularly misbehaving was leaving it alone in the dark for ~20 mins or so.

[@reilynn7891]

Yes, finally! Looking forward to the replicant uprising in 10 years, cheers guys!

[@0ptimal]

Man, I'm just over halfway through and have to stop and say this video is awesome. Bro what is this channel maybe I found a gold mine

[@potatoplanet14]

Grow a brain and right away torture it into being good at pong. Very uplifting for the future.

[@MK_ULTRA420]

China has been doing that for decades

[@jacobgoodrich6984]

we did the same thing to dogs, and other animals. When the animal does something we like we reward it, when it doesn't we punish. Same thing

[@potatoplanet14]

@@jacobgoodrich6984 It is not the same thing. Dogs can comprehend degrees of negative. A stern shout is less severe than a whack with a magazine etc. These "brains" can only comprehend inputs that make sense and ones that don't. Forcing it between its highest bliss and lowest tormemt.

[@FullMetalFeline]

@@jacobgoodrich6984 thats not good either lol

[@cheeeeezewizzz]

@@FullMetalFeline Sure it is. The training of animals like dogs has been for distinctly positive reasons. Training dogs to herd sheep is one particularly useful case of animal training. Training cows to remain within fences and to be, mostly, ok with human presence is another. Both have helped more people to live and survive. If we hadn't tamed dogs they would just be extinct like the wolves.

[@dax60840]

Born too early to explore space, too late to explore the world, just in time to be beaten at pong by your foreskin. What a time to be alive

[@takieddinbalti6956]

i wish to do undo this knowledge lmao

[@microska2656]

You can go to space, and you can explore the world. You're just broke and can't afford any of it

[@DJSekuHusky]

You could say it's on the "cutting edge" *I'll see myself out.*

[@stc.martin822]

I legit read that in 2 Minute Papers' voice. " Like I always say, one more paper down the line and we can teach foreskins to drive electric cars. Amazing."

[@gappuma7883]

Is the foreskin smart at all?

[@Freeksheu]

Man these videos are like fuel to my writer brain, lol. Giving me more ideas. Thx m8!

[@Sausager]

1965: in 2022 we will have flying cars! 2022: haha brain play pong

[@Mrsmifff]

I love how pong is the gold standard test for intelligence 😆

[@WhoThisMonkey]

Ikr, they should get it playing pin ball

[@_thresh_]

i think they should make the brains play super mario bros

[@MeatCatCheesyBlaster]

First pong, then breakout, then the world

[@awepossum1059]

Becuase it demonstrates simple thinking of future states as a result of performing an action, and also what angles are.

[@frostchain2362]

@@awepossum1059 True enough, I'd expect most mammals and reptiles and some insects to be able to play pong eventually with enough training.

[@khasanshadiyarov]

4:30, I am pretty sure, this little brain doesn't questioned itself about what am I or why am I doing the same work over and over again, because the pong game is all this brain ever received as an information about environment, so the pong is its whole world. Similar to us, we simply can not questioned about something we have never been affected by before.

[@akjohnny5997]

yea, it's not self-reflective

[@1010xxx33]

can't help but think about the upper beings with 22 senses sending stimulations to my brain so I can "see" or "feel" but really I'm just the equivalent of brain fragments on a plate playing pong and I'll never be able to reach reality, but I think you're probably right if you're incapable of sensing "the outside" it'll just be missing entirely in it's model of reality, playing pong or in my case try getting comfortable in the morning is really all there is.

[@blagoevski336]

Yup

[@nuassul]

Que buen comentario diste, es correcto lo que dices esos pequeños "seres" no conocen más allá de lo que le muestran los estímulos que experimentan con un juego de pong.

[@00MaTTaTtack00]

bro straight up this is like one of the most thought provoking philosophies. i don’t remember who but i’m sure you’ve heard of the shadows in the cave or however you put it. basically it’s about people who have been chained up in a cave and only able to see one cave wall. behind them is the whole world happening but all they have ever known is the shadows in front of them. that is their whole world. we have no idea what is beyond our reality. we could be just like the people in the cave, or these neurons in a dish, and we would never know.

[@WeirdBrainGoo]

I teared up when you said 'brain'. This is the most beautiful masterpiece of a video ever created. This is the most underrated channel and it deserves a Grammy for this video alone. I cannot thank you enough, I am crying as I write this and I haven't even watched beyond the point where you said 'brain'. Brilliant.

[@nskachuful]

Ar-.. are you one of those lab brains...?

[@SkyTheGuy8]

​@@nskachuful when the video is predictable

[@williamtoner8674]

thanks for explaining this so well

[@leandrobenitez292]

growing a brain to the size of a room... yeah that definitely won't turn into a survival horror final boss super villain

[@idk-zi3gw]

It has access to internet it's bout to hack everything

[@Obiwan7100]

Metroid's Mother Brain

[@cvdinjapan7935]

5:34 "The last thing we want to do is create systems that experience pain and suffering." He said it with a smile. Duper's delight.

[@stevensteven3417]

i saw that too.

[@stevensteven3417]

you are a good observer.

[@MAP233224]

Yeah there's definitely no other reason to smile while saying it, really none 🙄

[@akjohnny5997]

its tough cause without pain and suffering there's no motivation to improve, tricky.

[@cvdinjapan7935]

@@akjohnny5997 And it's clear that he knows that.

[@TristanMac65]

It's definitely experience existential dread when it misses the ball.

[@raptorhunter5549]

Imagine having a bunch of these brain dishes in a game like gta we'd be able to find out quickly what they're level of awareness is and maybe even be able to see how they react to each other

[@wooblydooblygod3857]

Well it would be extremely complex for them to learn so it would just be slight movements and occasional tapping for a very long time, they might never even find the reward or punishments so they would never have the drive to actually do something

[@award3007]

Let's see if they would get the low honor or high honor ending in RDR2!

[@olivergane8160]

Honestly wasn't expecting much from a smaller channel but great content! Pretty professional editing and presentation

[@ihmcurious]

Thanks homie

[@Gvozd111]

Everyone starts from the same place

[@billwaterson9492]

I've been going the other way on YT lately. If they've got more than 100k subs I'm like "meh".

[@sfurules]

Assuming that anyone worth watching has already been found is folly in my opinion. I've watched a couple channels now go from 100's to now 750K or so...all the while they just get better and better at it.

[@billwaterson9492]

@@sfurules I agree with that for the most part. A good channel going bad is relatively uncommon. I think the distinguishing factor between the two is the limitations of censorship accompanying notoriety.

[@gz6963]

Imagine being one of the brains born to play pong

[@stevensteven3417]

you basically are.

[@ptb2055]

i already am

[@TrueNativeScot]

imagine learning that you were made from baby foreskins

[@jonathanbarkins8480]

Says the one born to use excel 8 hours a day

[@InsertWaffle]

Born to shit force to wipe

[@hopo2hopo]

cool channel. subbed !!

[@mattongbp]

Good explanation!

[@cronchulus5489]

A brain grown on a chip, can’t see this going wrong

[@hipjoeroflmto4764]

This is how we get scorn

[@quixotes4478]

丂匚ㄖ尺几

[@Blameless902]

@@quixotes4478 Bro the tramslation

[@not888f9]

@@Blameless902 so many feet..

[@darthplagueis3488]

“Good soldiers follow orders”

[@shade7367]

"Researchers at UC San Diego are going to put human and neanderthal neurons into crab-walking robots and race 'em" This will live in my head forever now

[@Ocelot2009]

Go sapiens go!! Go sapiens go!!!

[@xoferwalken]

I want the scientists to yell let it rip like in beyblade before they turn the crab bots loose.

[@MogofWar]

Then the robots start fighting instead of racing....

[@willow_the_bob4322]

I thought this was a joke until I heard the narrator actually say this out loud

[@daviddaugherty2144]

I absolutely lost it when he said that, haha

[@blu12gaming44]

My greatest concern regarding these technologies is the risk of people devaluing life after it gets packaged into an everyday appliance or some other soulless product. When you become surrounded by things used as tools for everyday tasks: you associate their value with the tasks or products they are involved in. So what will happen when life is turned into a product? Eventually life may become so devalued that rights and dignity no longer mean anything.

[@pupsap7714]

So true

[@dramalexi]

3:05 So that's why I''m addicted to LoL 😮‍💨

[@PlutozReal]

"We made a life and we taught it to fear us." That's what I'm getting from this.

[@jdoe2737]

@xPyth How do you know? Even in the video they say that we don't know what it could be experiencing.

[@WackyrDrago]

​@@jdoe2737it thinks hit ball mmmmmm

[@MeatCatCheesyBlaster]

Training our first AI with child abuse. What could possibly go wrong

[@effervescentrelief]

It's the most human way to do things. Fear.

[@ViraL_FootprinT.ex.e]

​@@MeatCatCheesyBlaster They won't be happy until some paperclip making sentient super computer has destroyed us all.

[@christopherbeddoe406]

I feel like we are scratching at the outside of pandora's box from multiple directions and its only a matter of time before someone accidentally breaks through.

[@aspiretobe7320]

And we’re gonna wish we didn’t

[@volcom05345]

Like how we used the Large Hadron Collider to pierce a hole into an alternate reality which caused some of its reality to spill into ours causing the Mandela Effect.

[@toxicgracie3772]

@@volcom05345 🤪

[@flosa1995]

@@volcom05345 the alternate reality where you are tucked in the head a bit?

[@volcom05345]

@@flosa1995 I was waiting for a comment like this haha

[@mickmickymick6927]

Congrats on making it

[@devilmayclarify666]

YES! May this researxh continue

[@jaydenvengeance]

"Cortical labs is working on connecting the brains to the internet so people can program their own tasks for them." "You could grow a brain the size of a room!" Stunningly dystopian.

[@MichaelRodriguez-zr2ht]

Don't worry, if it progresses the same was computers did, they will become smaller and more powerful over time. Soon you'll have cyborg brains doing data processing, engineering, practicing law, etc. Corporations will have server rooms full of brains ready to outsource their computational power to any tasks asked of them. This will eliminate the majority of desk jobs.

[@jaydenvengeance]

@@MichaelRodriguez-zr2ht that’s not what I’m worried about

[@eggshapedisraelioperative6317]

Unless that brain could develop the way that humans do when they’re born, it would be completely retarded and probably only slightly more intelligent than the average American

[@Telados]

Can I grow a Tay?

[@dani.2479]

@@MichaelRodriguez-zr2ht Cyberdeck 2.0 (reference to cyberpunk 2077)

[@seraaron]

I think this is why it's good to keep the ideas of sentience and sapience separate. Imo, sentience just means 'capable of sensing and responding', and has nothing to do with consciousness. So by that definition practically all life and many robots we have today are sentient. Sapience is the far more difficult to define, and I would say that true sapience requires some degree of free will.

[@billymonday8388]

Im sorry, but how do you know that the dish cant comprehend himself? how else would it choose to pong the ball to get a reward?

[@kingdededelicious]

@@billymonday8388 those are the real questions

[@wesleydamen2018]

@@billymonday8388 its what you define as comprehend , it gets it stimuli from a machine and has no other way to get it , it has no senses no touch or sight ect. So the only thing interacting with it in a way it can sense is the electrical stimuli they give it so its not comorehending itself but it making snese of the inputs it has ordening it it knows what to choose because it likes order like most things in nature do so it chooses the path of most order and actively avoid chaos

[@billymonday8388]

@@wesleydamen2018 I do not define the words I use. clearly, the dish must be able to make sense of the inputs it produces. That way it can choose orderly actions more efficiently than if it was unable to understand itself. This has the consequence of the dish being aware of itself, even if its a low level awareness its still there. Of course the dish will never reach the self-awareness necessary to realize it is formed of neurones like we do. But then again, we dont understand higher level of awareness in a similar way.

[@jet100a]

@@billymonday8388 I'm not really sure about that. AI programs are rewarded or not rewarded all the time. It's not like a reward and that it feels good or that it makes you happy it's reward and that by changing a value or giving it a certain kind of stimulation you cause the behavior to repeat. Reward is probably not a great word to use because they don't mean it in the sense that people in their day-to-day life would use it. Unfortunately all I know about this so far is really what I've learned in this video and a couple of headlines that I've seen so I can't say this for sure but this video seemed to suggest that people are worried that the organic brain might actually experience reward as a reward in day-to-day life but it could just experience the reward in the same way that we "reward" an AI. One example of a way that you could think about a reward is that it could just be a reflex to a certain kind of stimulation or action. I think about the reflex that occurs when you're hit in your knee and your leg goes up. Or any other kind of automatic instinctual action. In other words it could just be that these neurons instinctially seek out repetitive understandable behavior so that they can survive but not with any actual intention to survive. When you look at animals and whatnot they typically try to be around things that they understand because something they don't understand is more likely to hurt them. Hell humans even have this behavior in the fear of the unknown and I'm sure all other kinds of areas. Sorry for any spelling or grammar mistakes or any strange word usage I'm using speech to text because it would take me all day to write this all out.

[@Casualbystander]

4:46 “We don’t know if computers can become conscious, but we know brains can.” Wow, I’ve never viewed the potential application of actual in vitro neuronal studies to influence AI or exist as its own network in a pursuit to exhibit consciousness… It’s true, we know the basis of consciousness derives entirely from our brain; the answer is there. What an amazing video! Thank you so much for breaking down this article.

[@memorablegan7920]

This guy always makes ML so fun to watch

[@Nik-dz1yc]

3:20 lmao I really love this video btw. I had no clue that our brains naturally aim to create predictable patterns and this makes quite more sense to be honest, although im still super curious to eventually learn the details of how neurons work

[@alf3071]

i guess if something is predictable then it can be controlled

[@ps5426]

“Scientists are actually preoccupied with accomplishment. So they are focused on whether they can do something. They never stop to ask if they should do something.” - Dr. Ian Malcom

[@leeperkillz5397]

Morality limits exploration scientists philosophy is to observe and keep observing in explore and keep exploring

[@Luftbubblan]

Awesome. Not unexpected at all. Will be interesting to see where this goes.

[@selfactualizer2099]

Like all things, experience exists on a spectrum. You cannot perceive the mind of something else, you cannot feel exactly how someone else feels. It's alive, and it wants to live, regardless of what it can or cannot comprehend. It does not have organs, or ideas, or anything else. But it is alive. We are not any different from a cell, A cell wants to live, a cell wants to reproduce, consume, grow, just like us. A neural network is a system of cells communicating.

[@AAjax]

I love that you worked with the definition of "sentient" rather than just pretending it's synonymous with human level AGI. The lamda thing was full of journalists, computer scientists, and tech company spokespeople asserting that lamda wasn't possible sentient. Meanwhile It's commonly held that fish are sentient.

[@mrkiky]

Yea but fish are considered sentient to a lower degree than Lamda. Lamda was being compared to human level of intelligence, which is nonsense.

[@ttt5020]

Many people think of the word “Sapient”’s definition when they hear “Sentient”

[@catpoke9557]

@@mrkiky Fish are much smarter than this brain.

[@mrkiky]

@Abraham Johnathan Yea I definitely don't deny that fish could be sentient and self aware. I think Lamda is not only below human level but below fish level as well. It doesn't understand what it says, it just spits it out.

[@AAjax]

@@mrkiky I agree lamda almost certainly doesn't understand what it's spitting out, any more than a fish understands primal urges that lead to more complex behavior; the fish doesn't understand it's creating a new generation of fish when it drops eggs, or fertilizes dropped eggs. My issue is people dismissing *any* level of sentience on the basis that lamda isn't self-aware. Nobody knows how sentience emerges, but they confidently declare that lamda can't have it. Sheer hubris. It's possible to be cruel to sentient beings, even if they aren't particularly self aware. We're setting the stage for exactly that.

[@gfullcrayon562]

Imagine getting the short stick in life and having to be a mini brain that perpetually plays -pong 😂

[@ima_doot7827]

So no taxes, no job, no school, no worrying about the future, no social anxiety and no responsibilities? I feel like we got the short stick in life

[@gfullcrayon562]

No one stopping you from being a bum

[@satabell5551]

@@ima_doot7827 You're right. :put gun to mouth: Time to reset

[@FractalNinja]

I think teaching it "random" stimulus is "bad" and "organized, predictable stimus" is "good" could go terribly wrong. What happens when a cyborg decides humanity is too unpredictable to be pleasant, and decides that the only way for every stimulus to be organized and predictable is if it stops humanity 😮

[@michaelshada262]

HAY! You're talking about every gobber now that lives in their mon's basement playing games thinking someday they will make millions at it! When very few make a dime.😂

[@scottie1858]

I see the train to horrors beyond imagining has left the station and is now out of control...well done humanity, well done!

[@redhandedninjastar]

This is just phenomenal

[@bansoma]

If we have to live in a dystopia, why are we constantly seeking the worst possible version? Imagine how you would feel to be the brain-jar-person this is leading up to?

[@stagnant-name5851]

All you would feel is a drive not be punished and a drive to be given a reward. No emotions No personality.

[@samusaran7317]

@@stagnant-name5851 People are already heading in that direction for quite some time.

[@gabemerritt3139]

@@stagnant-name5851 So intense slavery combined with sensory deprivation and social isolation?

[@gabemerritt3139]

Well imagine Plato's Cave. These brain Jar people aren't going to know the world except through their own eyes. How do you like being locked in your meat suit, forced to work for food, water, and shelter? Would you even want to give up the devil you know for something unknown?

[@HavianEla]

That’s what I’m thinking. I’ve never actually played the game, but it reminds me SO MUCH of SOMA. This is REALLY scary. Just because we CAN do something, doesn’t mean we should.

[@kirby89000]

Imagine your whole existence is playing pong at a pro level

[@randyg666]

I feel like that sometimes

[@Blox117]

imagine being created from a foreskin

[@golevka]

Forrest Gump brain

[@Fence07]

What is my purpose? You play pong.

[@EdgarMartinez-ws9ws]

This is a wonderful video. Now, I want to become a Neuroscientist.

[@HonestGraduate]

The amount of information in this video is outstanding! I appreciate you share the name of the studies.

[@yourmom-in4po]

This has been the best and craziest video i've seen this year, this is truly a masterpiece, thanks for sharing this with the world! really deep stuff

[@pattyihm8099]

I am actually Elliott’s mom-love your name🙂. Elliott ate a glow-in-the-dark star as a baby. I have always believed that it added to the depth of what’s in his brain.

[@danielash1704]

In a way I'd say they are a common intelligent plasma reflection charges v building upon the compressions to expansion of sensory information about subsection to opperation as a cell biology there is a big deal with the source center of frequencies braiding and memory pathways in the plasma glucose tolerance for loose ends unfolding information into us from the outer edges and memory of the experience to react.

[@aeonremnant]

I honestly love progress like this. :) So awesome.

[@Kaasbaas045]

Great huh, you people have no idea how demonic this is. Society is so out of touch it is mindboggling. But I geuss people like you will never realise how evil this is. We are losing our connection to the soul.

[@AB-wf8ek]

"The last thing we want to do is create systems that experience pain & suffering" - sounds like an oxymoron when the system they're creating learns by punishment. We shouldn't shy away from systems that experience pain, since it's integral to the learning process, what we should be developing is a definition and implementation of compassion in tandem with the experience of pain.

[@cringy8095]

Until it is experiencemented on you

[@turolretar]

That conclusion could be taken outta context

[@mrkiky]

@@cringy8095 We all experience pain and suffering. It's part of life. That's pretty much how you teach your kids to be part of society.

[@mr.skeleton5536]

@@mrkiky With pain and suffering. There's another way, Satan.

[@mrkiky]

@@mr.skeleton5536 No there isn't.

[@samarthtripathi8397]

You have such a narrator voice, could have listened to it for hours. Great video man!

[@Mornings]

Imagine dying in CS and you see in the chat "oi mate ye just got one tapped by a fookin brain in a jar!" 💀💀💀

[@chadthaddison]

Great stuff! Keep this up and your channel will blow up in no time.

[@pyrrhusofepirus8491]

We may have created what just, even slightly be the first artificially created sentience. And the first thing we did was make it into a gamer… All I can say, is even if he’s just a dish brain, he’s one of the boys

[@smolchungus5647]

dish brain sounds like a really original insult

[@diablo.the.cheater]

@@smolchungus5647 pass 100 years and it may be considered a slur

[@alexamderhamiltom5238]

he is indeed, our kind.

[@xexzersy]

@@diablo.the.cheater XD

[@BeesKneesBenjamin]

About 2 decades ago they were using neurons grown in dishes to stabilize and fly planes in a simulation...

[@genericname2284]

I have a Petri dish locked in my basement rn. I’m just constantly feeding it noise and monitoring it’s suffering via an arduino attached to a tickle me Elmo

[@randospawn7495]

Theses concepts are cool especially the idea of creating sentience but obvious ethics are at hand and if they can experience sulfuring and confusion tests should be minimal also if they can data storage may be greatly enhanced with these and creating a system they can learn off of and display info from in human mannerisms would also be interesting

[@Triisttan]

Super interesting video as a highschool Senior planning to major in Computer Science with a minor in AI next year. Pretty impressive that you're publishing videos of this quality as a smaller youtuber, I hope you make it big!

[@BayLeafff]

Amazing presentation man! Was honestly surprised to see that this came from a

[@anhtqyrc]

To be able to know if a being is conscious or not, imo we need to see if they can process abstract information, e.g forming beliefs, generalizing,... Which are hard to quantify and test.

[@goldsilvervscrisiscollapse4320]

"what is my purpose" To play pong "Oh my God"

[@0xC47P1C3]

Your writing is excellent for these topics. Subscribed!

[@MeinCouch123]

okay but what the hell with the foreskin thing, is that why circumcision is so encouraged? It's literally free money for the hospitals, and the parents aren't getting a cut (well, the baby is)

[@mattlm64]

Yes, it's baby body part harvesting for profit.

[@_hurglet]

If they develop these guys into full beings then they stand no chance in a roast battle.. "You're calling ME stupid? Your brain is literally grown from foreskin."

[@ludologian]

My question is that does the brain cells see the direction of that ball? can it actually "see" or what kind of input it receive to know where it's going? calculation ? second question is; - about it's actions inherited from what system? can we predict if it's from system 1 or system 2? only some system biologist/ cybernetican can answer this question? if these cells react to simulant through predicable environment. it shows some interactome but can this emerge into not just learning but infere behavior? or ut can mix between the two? lastly if this somehow remember some minimal informations can we actually transfer these or reuse it for other cells just like in computer where we can save a learning model weights into serializable format

[@MegaShrooom]

Really interesting video. Appreciate how you dived into the question of consciousness. Glad youtube is recommending me smaller channels more regularly.

[@CameronfDrums]

this is an incredible find, and very well executed video. Thanks for this.

[@vardus7930]

This is horrifying. Imagine being so bad that you will reincarnate as DishBrain and you need to play this fucking pong for eternity.

[@homelessrobot]

"Like a baby learning to use a fork" > baby twirls a spoon between its toes, farts, and makes itself cry

[@tuckerfrancis2828]

Even single celled organism exhibit signs of sentience by adapting to novel stimuli, even stimuli that would not be found in nature and therefore not specifically evolved for. Seemingly, they do this through problem solving, often in ways that is beyond trial and error (which would be expected if there were no sentience) I think basic sentience emerges at the cellular level and the intracellular communications just organize the sentience of each individual into a hierarchy of increasing complexity, giving rise to the emergence of abilities that would not be achievable by the individual. Each cell contains the ability to store information endogenously... But, only when organized into a network, can they begin to store information exogenously. Animals(typically) are able to store information endogenously, but not exogenously through symbolisms. Humans gained the ability to store information exogenously through complex symbolism, and therefore we were able to communicate complex ideas reliably. This collectively amplifies our consciousness, and ability, to a level that would not be possible otherwise. Neurons increase their sentience collectively in the exact same manner. While each individual neuron is sentient to some extent, the collective sharing of exogenous information gives rise to a higher consciousness. The mushrooms tell me that this is true. As within, so without.

[@ReasonMakes]

I'm not convinced that a city or website is sentient. I just don't see there being any cohesive subject experiencing life there. Similarly with cells. The simplest I've seen sentience emerge is in insects like bees, who are certainly sentient with emotions and even their own language.

[@andrewjackson3959]

Animals store information through smell. That's how wolves know exactly where their territorial borders are, for example.

[@tonoornottono]

@@andrewjackson3959 lmao

[@flambambam3578]

It gets even crazier when you think about how biology is just chemistry at the smallest scales. "Sentience" and its emergent properties are really all just a bunch of atoms interacting to reach stable states. I just think that its so fascinating that simple physical laws (electromagnetism in the case of atomic chemistry) are able to give rise to systems complex enough to "understand" themselves.

[@peterjohnson11655]

i think the simplest answer is that concepts like "sentience" and "life" don't really have any definite meanings. there isn't any law of the universe that makes some things conscious and others not, if you look deep enough into any living being, we're all made of the same fundamental particles that everything else in the universe it. there's nothing differentiating living, breathing beings from other matter; it's not like there's any way to quantify consciousness. is artificial intelligence sentient? are cells sentient? are cities and countries sentient? i think the true answer is that sentience itself is just a concept made up by our brains, and as such, there are no real borders between what is and isn't alive, and trying to find that line is a fruitless task. i think that instead of trying to simply categorize things as either "conscious" or "not conscious", we should look at it as a large spectrum where some things are more conscious/aware of their surroundings than others. thats just my perspective, though. im not like a philosopher or a scientist or anything like that, so im not an expert on this subject, but this is just what makes the most sense to me.

[@digitalclown2008]

As long as it's cells in a dish, i think the ethics don't matter as much. But if it get to the point that there are more sophisticated organisms being created, at that point it gets more complicated. But I'd be lying if i didn't say I wanna see how far this can go.

[@Alex-wg1mb]

Yep. It is like in a Bladerunner movie. Synthetic humans that want to be free and alive

[@tonoornottono]

@@snuurferalangur4357 straight up didn’t watch the movie then. it doesn’t matter why we built them if they feel, if they exist consciously. feeling, conscious beings deserve liberation.

[@yeehmm1893]

vegooning

[@foppypoof5195]

@@snuurferalangur4357 How sociopathic of you. While I don’t believe we can create sentient artificial life, if they do end up being sentient, and have emotions like us, then they deserve to be treated like human beings.

[@TheJunky228]

the ethics of getting those cells are absurd though. circumcision is genital mutilation and the stolen tissue is sold for profit in this case

[@austinrussell101]

This plus a neuralink could possibly mean that we could have a backup to our brain info i find it very cool that i might see it come to life in my lifetime since im only 21

[@droid16beta97]

Good video.

[@PSTMRTMe]

Hope this reaches the Algo, it's a really interesting topic and good quality production.

[@larion2336]

Good video, really interesting imo. It almost seems like maybe neurons have evolved in a way where their reward mechanism is to detect new patterns and adapt to seeking that out again next time. It's a very simple idea but seems like it could be at the root of much of intelligence as we know it. I'm reminded of the neural nets that were improved significantly in navigating maze like games by introducing a curiosity reward function - they were rewarded for seeking out novel stimulation (in the form of frames/images of the game not seen before) and therefore they developed a kind of memory since the areas they had been to before would be less reward than going to new areas. One fly in the ointment there was that when a TV showing random images was introduced to one of the walls in the maze, the AI got completely stuck on watching the TV because it was always showing new images, so it stopped exploring the maze altogether (which really has a lot of real world implications for people IRL being distracted by movies/games/TV tbh lol).

[@ihmcurious]

How human of them! I hadn't heard of that study, super cool

[@mrkiky]

AI: I am in pursuit of happiness Drugs: *gets introduced* AI: hmm, what do we have here?...

[@AnimeG_irl]

I can't believe they gave an AI ADHD

[@KoDaring]

absolutely correct about the internet and the overstimulation

[@catpoke9557]

Also kind of terrifying because it shows that the desire to be stimulated goes beyond an emotional level. The neurons themselves are craving it and driving you to seek it out. It really helps you to understand why things like addiction are so hard to break out of, or why things can be so difficult to do when stimulation is low.

[@jasonstruthchannelsearchin2670]

Meshuggah - they move below video was uploaded 4 days ago and it was under this video with the algorithm. It looks like a thinking matrix of scary stuff similar to this video uploaded 2 weeks ago.

[@GeoffRogers42]

Oh god that poor little nugget of sentience, trapped in pong world with no way out

[@eddielopez2373]

Right out of the gate: “it sucks at pong!” 😂 maybe this is how an advanced species is analyzing our behavior from a different planet.

[@zombiedemon1762]

Aliens: Sometimes we wonder why we don't visit humans but then we remember that they suck at Pong.

[@zombiedemon1762]

@@wwrrrrr3579 . ?????

[@joan_serrano]

I still wonder why neurons prefer 'predictable patterns'. Could this preference be linked to the random stimulation that causes some neurons to become excessively active, eventually resulting in excitotoxicity and their death? Surely, there must be a simpler biological explanation for this phenomenon. Any suggestions or references? Thank you!

[@geraldkenneth119]

A comforting thought is that, even if DishBrain and similar neuron cultures are sentient in some way, odds are they don’t experience emotions since they don’t have a Limbic system. They might not even experience pleasure and pain (at least in a way we could understand) since they don’t have pain processing circuitry or a reward center.

[@sebastianjost]

Very interesting video. I'm glad the algorithm recommended it. learning this much in 15 epochs is incredible compared to reinforcement learning! One big improvement you could make: Add references in the description (especially link the paper you talked about). I would love to read more about this.

[@ihmcurious]

Thanks for the tip! When I added links to the description the views dropped like a rock, so I'm trying to figure out if I get penalized for sending people off KZbin. But here is the full DishBrain paper: www.cell.com/neuron/fulltext/S0896-6273(22)00806-6

[@sebastianjost]

@@journey8533 seriously? That sucks... But sadly it's perfectly believable as the algorithm is supposed to maximize time on KZbin.

[@neoqwerty]

@@journey8533 Pinning it still works as a trick AFAI've seen-- though placing it as a reply to a comment asking for th source is probably the best way to get around the algo for a bibliography, still.

[@robertc.5558]

@@ihmcurious put links in a pinned comment.

[@glory8864]

15 whats

[@zefellowbud5970]

Ive had a similar theory in the past that assumed pain as a chaotic randomized signal. Heard a while back that apparently people who experience chronic pains experience difficulties with learning and remembering, not to an extreme extent. Honestly mostly based on hearsay knowledge from my end, but its nice seeing others also has had similar thoughts, neat

[@jesper9622]

Pain is probably not only chaotic and random, but also very strong in magnitude compared to normal stimulation. It makes sense that a big pain will cause big signals and many neurons to be affected by the signal.

[@zefellowbud5970]

@@jesper9622 oh yeah better yet theres probably varying intensities. From dull pains, sharp pains, to incredibly burning pains.

[@official-obama]

@@zefellowbud5970 some ouchie points over here, ouchie points on this line, and a _bunch_ of ouchie points over here edit: aaggh no

[@zefellowbud5970]

@@official-obama lol The convo makes me go If chaos is pain Order is a reward Does that make us innately obsessive compulsive

[@jesper9622]

@@zefellowbud5970 yeah and the punishment can be calculated by the area under the graph P = h * t, where h is the magnitude and t is the time that the pain is applied. A dull pain could be applied in the scenario when an agent is simulated to go around a maze and eat food. It has a hunger level between 0 and 100, (100 is full stomach), and each moment the magnitude of the pain is equal to some constant k*(100-hungerLevel). This dull pain will help the the agent to learn faster compared to a sudden shock of pain at 0 hungerLevel. In A.I. literature this is called dense rewards and sparse rewards.

[@abhmmh8892]

"what is my purpose" "You play pong" "Oh my God"

[@HansDunkelberg1]

The crucial question remains unanswered what mechanism triggers the predilection for the more predictable response. Comparing the reaction to the one of a human being doesn't help.

[@prosamis]

This is so trippy For just CELLS to have "preference" to more order is so crazy to me

[@reginaldthunderfuck4832]

This just sent the sci-fi reader part of my brain absolutely nuts. Awesome video.