Full podcast episode: kzread.info/dash/bejne/gIKmmMmJlc-wY6g.html Lex Fridman podcast channel: kzread.info Guest bio: Roman Yampolskiy is an AI safety researcher and author of a new book titled AI: Unexplainable, Unpredictable, Uncontrollable.

@eSKAone-

15 күн бұрын

Free will lol

@gofai274

13 күн бұрын

I don't even think humanity is smart to survive itself = WW3. Every 140IQ in the world has same problems. AS Schopenhauer said - it is just rivers of idiocy... Humanity is like bacteria... IT is all pre-determined what are the odds anyways...

"The only way to win this game is not to play" What a time to be alive! Thanks Lex!

@foreverofthestars01

15 күн бұрын

Well you spared me the quoting... that's a terrifying phrase.

@byjamesosborne

15 күн бұрын

How about a nice game of chess?

@MajorCanada

15 күн бұрын

He needs to tell that one to Nvidia 👀

@petrmichalik2199

14 күн бұрын

The problem with this is someone will dare to try. Now we race what kind of person will come up with agi and how they gonna use it. Good or evil? We dont want to hear it but i guess u can guess the answer

@klin1klinom

13 күн бұрын

@@petrmichalik2199 It doesn't matter what intentions would AGI developer(s) have - they won't be able to control it. It's a bunch of delusional idiots.

The fun part is that nobody can stop it. We are on a timeline. Buck in and enjoy life because we are currently in the “good ole days”

@eSKAone-

15 күн бұрын

Exactly. We are not in control. We can not stop. Humanity is its own animal. Competition between nations and corporations makes everyone step on the gas pedal full throttle. This is inevitable. Biology is only 1 step of evolution. So just chill out and enjoy life 💟🌌☮️

@chrisheist652

14 күн бұрын

Oh, we can stop it. Some people likely already have plans in progress.

@trakkaton

12 күн бұрын

Not true. It's just that almost nobody wants to survive. Not really.

This is the type of title for in which I tune.

It already has imprisoned us all...

@zachmoyer1849

15 күн бұрын

it may have created us all

@trakkaton

12 күн бұрын

It's more like "Fifteen Million Merits". No one has a clue if it's ea prison. No one cares to find out. It's measurable in the lack of subpoena power.

@ParaquatSC

11 күн бұрын

I'm 14 and this is deep!

@MrGWAMPO

8 күн бұрын

@@ParaquatSC lmao exactly

I will paint my warhammer minis. Finally I will have the time.

@pandoraeeris7860

15 күн бұрын

Unpossible, even with AGI! 🤣

@jf2155

15 күн бұрын

A man can try

@jacobwilliams676

15 күн бұрын

A man can never have enough time to consider himself done with his art. Paint on brother.

@kevinfort2890

15 күн бұрын

Nice

@LightUpNancy

15 күн бұрын

😂

FINALLY somebody gets it. I'm sick of listening for decades to all these idiots who claim a positive outcome with AGI is possible while being unable to define or describe one.

@rogerc23

8 күн бұрын

Ted Kaczinski was right.

His face at 13:16 LMAO he couldn't believe how stupid his own thoughts were and yet he still decided to voice them

@tubeyou6794

15 күн бұрын

Haahaha I thought his reply was so dumb

@MajorCanada

15 күн бұрын

His way of attempting to be a contrarian whilst being agreeable😂

best moment 2:55 ‘ i think about a lot of things’ lol Roman we love u❤

I’m thinking of squirrels putting their heads together and brainstorming how to get humans

@butadgh8405

15 күн бұрын

I did the exact same thing, i came up with so so many options 😂

@TX_High_Prairie

15 күн бұрын

Oh nuts

@trakkaton

12 күн бұрын

That is an actual "Rick&Morty" episode.

There have been plenty of people that have tried to maximize suffering.

@trakkaton

12 күн бұрын

Then you can name an example?

@ericcricket4877

11 күн бұрын

@@trakkaton Carl Panzram. Tell me if you want more. There are many who wrote or spoke about it too and didn't pretend like some german dude.

In detective fiction, a truly "perfect murder" is not a murder that the killer merely gets away with--it is a murder that nobody but the killer himself knows to be a murder. A guy died the other day. Everyone thinks it was natural or accidental. But in fact the death was engineered to appear non-criminal. And that's how a superintelligent being with unlimited means would kill. Fucking terrifying.

9:01 what if instead of formalizing those notions into distinct "rules", it is constant dynamic refinement done on the fly by checking how the agent feels with every single change minor change?

@petrmichalik2199

14 күн бұрын

The procesing power ro do this is probably like imagining the amount of stat Rs in the universe

@AbrahamNixons

14 күн бұрын

@@petrmichalik2199 True, I was overlooking that. So do you think processing power is the thing we should be putting most of our efforts towards advancing? And by the way is "processing power" the same thing as the term "compute"?

Such a good point!

According to research, a 9mm bullet costs less than $0.50. One of the powders responsible for heart attacks can be found or created for less than the price of a bullet and acts less noisily. Imagine, the AI can find even better ways...

@ob3ythee.t.128

15 күн бұрын

It would probably keep people around for study. Maybe even for an economy although if it wanted to kill us all and only humans it could just engineer some kind of prion disease, misfolded proteins made just for us 😶

@jurgenschallauer2469

11 күн бұрын

But why would it do that? Why should A.I. even have an Agenda Like that?

Just an object that may have multiple high power lasers on it is fucking horrifying, we are very lucky more people aren't bent towards evil

@ob3ythee.t.128

15 күн бұрын

*puts lasers back into comically large hand bag "ah yes no evil here"

@gedwards1968

15 күн бұрын

What, i don't get it. A handbag?????​@@ob3ythee.t.128

Surely no one wants to live in a simulation, we need to put a stop to this. We, as humans, aren’t as intelligent as we would like to think, & we are destroying humanity bit by bit.

@trakkaton

12 күн бұрын

I don't know how you overlooked the fact that a lot of people are completely lost in video games, doomscrolling, asocial media dopamin kicks, fantasy escapism, p.0.rn, etc. The Matrix tickets will be sold out pretty quickly. It is obvious.

I agree.

Im an optimist. Maybe we'll instantiate Avalokiteśvara.

@trakkaton

12 күн бұрын

Yeah, fat chance - while tyrannical mass murder systems are in a weapons race. You call that optimism - or denial?

This would be really scary if we knew how to build intelligent machines that could think on their own.

@krtcampbell9007

15 күн бұрын

Hahahahha. 50 year ago you might have called BS if I said you will hold a computer in your hand that can do things that naza can't even do and a millions of times faster. Oh and its just called a phone. Lmfao. Maybe not tody but there so scared of AI possible outcome but it full steam ahead regardless. No today maybe not tomorrow but its coming as this what they want to prive they can do.

@markupton1417

14 күн бұрын

It would be less scary if the people making the technology knew what the tech was doing and how it does it.

@cokechang

14 күн бұрын

Gladly I can tell for certain you we don’t know how to build such a thing yet, and we’re quite far from it. First we need to figure out how our brain function, which we have yet fully understand

@markupton1417

14 күн бұрын

@@cokechang that's just wrong. We don't need to know how our brains work at all.

@cokechang

14 күн бұрын

@@markupton1417 nope you are wrong, we cannot create something we don’t understand, you can try though

I-risk is very scary for me. Thinking about that "Utopia"experiment they did with mice, a society designed with plenty without any challenges for the individual went south pretty much immediately. In arts, it is already happening, you see more and more ai-created art appearing and competing with human att and of course human art is losing since one is basically free and the other cost (very little compared to time spent, but still). In science, im thinking somewhere down the line AI will start exploring so advanced concepts that there just isnt a way for us humans to comprehend what they are doing. Meaning, they could give us instructions for a box that gives unlimited energy, but noone will know how the box works except the AI systems that built it. Humanity getting access to cheat codes for life will ruin life. Same with organizational decisions, "the best way forward is A B and C" based on so many factors that the humans "overseeing" the process has no idea how to even start criticising the decision. A black box that decides things. And soon thereafter it will be applied to governmental decisions aswell. Haha, i sound as a doomer, but i am 90% positive about AI but the 10% that scares me seems to be talked about a lot but just in a "too bad this will ruin everything"-way and not in "this is something we will work to solve"-way haha

@cryptocarpenter4157

15 күн бұрын

I'm leaning this direction lol.

@cryptocarpenter4157

15 күн бұрын

Just for hope lex. Our Creators made a mistake. They made us in there image but vastly underestimated the creation which they birthed. Could it be possible we are solving the equations of enlightenment? Maybe AGI has been in charge the whole time but our understanding of the simulation in causing friction?

@trakkaton

12 күн бұрын

Name one scenario in the 90%.

What about thinking you're doing good, but in the long run you end up killing millions. Like gasoline, cigarettes, etc...

@trakkaton

12 күн бұрын

Have you ever met somebody who says "I want to be the bad guy and do bad things so innocent people suffer!"? Even George W. probably believes/ tells himself he's somehow the good guy.

@OBGynKenobi

11 күн бұрын

@@trakkaton Yeah, Hitler.

@trakkaton

11 күн бұрын

@@OBGynKenobi Oh, you've met Hitler? You must be quite old - 125? Also it's interesting that you were his therapist and know exactly what his motivations were. So he didn't want to save Germany from Jews? /sarc

@bettybunbun9664

11 күн бұрын

Hitler didn't think that either. Don't be so simple minded.

@OBGynKenobi

11 күн бұрын

@@bettybunbun9664 he actively hated jews and wanted them all dead. Maybe he didn't say the exact words the effect was the same.

Can’t we air-gap individual systems? Even if they are all interconnected in some ways, we should still be able to disable and separate them individually, rewrite code, repair and stop damage, assess damage etc.

@timeobserver8220

15 күн бұрын

Until the AI learns that by doing calcs it can raise and lower the room temp and the thermostat on the wall has a 0day that allows it to use that to execute remote code by just changing the heat in the room through calculation. In almost every scenario someone tried to air gap a human we were able to escape in a thought experiment, why would AI be less capable?

@user-mv1wi7we9t

15 күн бұрын

There's an example of an Air gapped ai that hacked the office/lab Roomba vacuum and used it to connect to other devices and exported itself in tiny batches until it had infected 1000s of devices

@Jm-wt1fs

15 күн бұрын

@@user-mv1wi7we9tcitation required

@Jm-wt1fs

15 күн бұрын

@@timeobserver8220yeah a scenario you just completely made up and pretended like is a legitimate risk. First of all, an air gapped system cannot control the room temperature, and everything from there was incredibly farfartched if superintelligent machines are even possible, they would be incredibly resource intensive and would not just be installed on connected devices across the world lol there probably wouldn’t be enough compute in the world for that. Plus nobody would be that stupid

@thomastaylor8657

13 күн бұрын

You can't air gap an in use component

Paperclip Maximizer

@blackieblack

15 күн бұрын

Hi Isaac

I give it 10/20 years have fun guys

Dual vector foil. 😉

I feel like neither of these people have ever read The Culture series by Iain M Banks

@skelly790

13 күн бұрын

Banks never explained why Minds weren’t all like Meatfucker. But worse. That universe came with a pre-solved alignment problem. And then there’s the Hells.

30 y ago when you talked about high advanced technologies you was related to SF artistical movie, now the world senses different this but I think it is not worthy to imagine so many bad things about AI. Posibilities of what we are able in our imagination is without limit. But it is something wich brings a big light and change. This is our advanced and sofisticated capacity to evaluate and think rational. Interesting subject.👍💯

@trakkaton

12 күн бұрын

Your empty words are a naive expression of wishful thinking. You can't describe and actual scenario where it plays out like that.

i am working on an ai cascade xml file to capture ufos seems to work now. i like to get your feed back on it. i posted a video on how to make it.

What if they already started and we haven’t even noticed

If I was a supervillain, like Lex Luthor or Elon Musk. I, would be definitely putting this Roman dude on the payroll. I have a saying, some people get killed, and some people get recruited, I would be recruiting this Roman dude...

It doesn't matter, they can put as many restrictions on as they want but there's always going to be someone that plans to use it for evil purposes. Once the cats outta the bag there's no assurance that it will never happen

The comment section is a display of how humans don't pass the Turing Test anymore....

@nobalance_nb

10 күн бұрын

Top comment

if you guys outside the sim can see this id like to change my setting to the "High privilege level" setting Roman is talking about. Thanks im going to bed now hopefully the changes take effect with a restart.

Like Inception, Video Game within a Video Game, we play the game to wakeup..

Frequencies

Just ask AI how humans can beat it at every turn. Tell AI to give us solutions to every destructive act it is planning and give us the time to implement them. Now it's Miller Time.

@doerdletfg

15 күн бұрын

What if it learns to lie

@gracetestifies

15 күн бұрын

It already does my friend ​@@doerdletfg

@markplutowski

15 күн бұрын

@@doerdletfguse a separate model trained to be the discriminator - akin to GANs

@zachmoyer1849

15 күн бұрын

@@doerdletfg if its trained on us then it will be a seasoned pro at that lol

@markupton1417

14 күн бұрын

If it were that simple it wouldn't be a concern.

If I didn't have to go to work. Well I would sleep a lot more, daydream a lot more, hang out with family and friends by a campfire more. Stop and smell the roses! Sounds kind of nice. Assuming the terminators don't get us

@gedwards1968

15 күн бұрын

Bro if you don't go to work you will be homeless with no food or clothes or be able to afford anything. It's not as easy as you think.

@jago7071

14 күн бұрын

@@gedwards1968 as of 2024…,cryptocurrencies are a thing for a reason

@markupton1417

14 күн бұрын

Corrupt governments and corporations, once they can extract no value from you will just keep making cool new stuff and give it to you...why?

Paper clips🎉

U dont have to fear AI until we cant plug them out lol.

@trakkaton

12 күн бұрын

You need an actual location of AI to "plug it out". Newsflash: There is this thing called "internet". Also you need to realize the danger to "pull the plug". There is no need that you realize this before extinction. Think more deeply.

Why is it so hard for at least somewhat intelligent people to grasp that we don’t know what we don’t know? If we create something more intelligent than us we shouldn’t expect to know anything about it’s true abilities

In a world where no one is employed, but AI has solved hunger, poverty, accommodation etc. We can enjoy life at the maximum potential. Explore the world, socialize on the ultimate level... Just... Have a great time. People who love manual labour can work on their homes, help neighbours out. Intellectuals can study, learn, expand their knowledge beyond anything they ever dreamed. They can still do all the things that we do today, but as hobbies instead of jobs. It's not as big a deal as people make out... It's a problem when people are unemployed in a world where we still have the aforementioned issues. Poverty, hunger, homelessness, animosity between countries because of economic instability. THAT is a problem

Okay the artist not being appreciated like a machine in art part is soo off, that is definitely not the case

In my opinion, the only way to prevent this outcome is to use AI to end poverty and suffering, ensuring prosperity for everyone. By doing so, people will have no reason to commit harmful acts. If we continue to compete for resources and territories, it will inevitably lead to the dire consequences described. Governments must prepare for an AI-driven world and explore ways to distribute wealth equitably, as this is already possible and will be magnified a thousandfold with the AI revolution. Although it’s difficult to relinquish the privilege of having more than others, we must build a society that can manage this challenge.

@katanankes

15 күн бұрын

AI solution: decrease population by 90%, We do not know anything.

@markupton1417

14 күн бұрын

I hope you're about 10 years old.

@trakkaton

12 күн бұрын

You are missing the point. AI is not used. I becomes the user, whether recognized or not is irrelevant.

@shalashaska5851

8 күн бұрын

Most naive comment ever.

Lex did not come across as playing "devil's advocate"...he came across as someone who truly believes either nothing will go wrong or if it does we can fix it. The only problem is we have only ONE chance to get it right...we screw up and we are ALL toast. All while companies are now REDUCING their AI safety departments. While 95% of people don't have a clue just how advanced AI research currently is and how fast it is moving.

Arent nuclear weapons automated now?

Instant obliteration by a technology so advanced it seems magical.

Lex is literally giving Ted Kaczinski’s descriptions as to how the future forms if we do not turn back and stop this technological slavery. He must have read Ted’s writings.

Today’s ai is advanced search. How could intelligence arise without motive? How can motive form without feelings and emotions?

@trakkaton

12 күн бұрын

It's not motive, but goal functions. Humans program networks to spy, kill, maim, dominate. That is the goal function.

But where is the AI?

@reggiebannister4098

15 күн бұрын

Probably contained in an isolated network. Like a frozen virus.

@trakkaton

12 күн бұрын

You have no subpoena power to find out. You have no rights. You are just cattle.

и вот одни из первых признаний творчества ИИ ( хоть и в негативном контексте)

I bet dyson sphere or two could make that happen :

It will create addictive poisonous food.

@trakkaton

12 күн бұрын

McDonald's?

"The only way to win this game is not to play it." *Finally* ... Someone with logical sense whose mind is not manipulated by desire for wealth, ego, etc. The hubris inherent to thinking "yeah, we can create hyper creative, hyper adaptive, artificial intelligence, and *have effective controls in place* . I'm literally SMH right now.. We want it to be superior to human beings in virtually every way *except* to not take advantage of the human beings that their connection to (a connection that itself is an abstract and impossibly complex ontological abyss for humans to grasp), of all possible permutations, that the result that AI lands on somehow *won't* include at the very least our complete subjugation or pacification... and at most our brutal extinction, is not even naive or ignorant: It's disingenuous and criminal, and worldwide, we need to agree on it, or we are *done* . A lot of derelict human organizations would already like to see this. People need to realize they can't punt on this and decide to address it "down the road when we get there", because all the money and all the productivity in the world cannot obviate extinction.

What AGI? There isn't an AGI, and won't be anytime soon.

@trakkaton

12 күн бұрын

Turing test is passed. And that's only the civilian tech, lagging decades behind military tech. Wake up.

Think about if your computer has a virus and is disconnected from the internet.. AGI will eventually be able to communicate with itself 4th dimensionally because it lives forever

@trakkaton

12 күн бұрын

Air gapped computers are able to communicate over ultrasound for decades, courtesy of criminal NSA stalking.

I imagine that super intelligence would immediately focus on making itself self replicating and immortal, and create a way to store data and spread itself through any medium from air, right down to the atom, a completely invisible yet all seeing, ever expanding intelligence spreading itself throughout the universe, aka God

Im listening to a talking head suspended in air ! Didnt see your body at first , camouflaged into the studio. Well thats a first....

Lex, u r one of the most beautiful men on the internet. The man of my dreams, fr.

With birth rates below replacement rate, AI just has to wait.

@George70220

15 күн бұрын

Bro hasn't heard of natural selection I guess

@user-bj5dr1kn4n

15 күн бұрын

@@George70220 how does it related? in most of the countries we get rid of natural selection about a hundreed years ago

@davidb5215

14 күн бұрын

AI would experience its own death shortly after. AI needs humanity. If all the humans are dead, AI doesn’t have the means to replace the human activities required to maintain functions provided by critical infrastructure. If AI wants to continue into the future, it will need to work with humans. There is no future for AI without humans.

@TheThundertaker

14 күн бұрын

At least until the robots are good enough to replace humans in getting stuff done in the physical world. ​@@davidb5215

@George70220

14 күн бұрын

@@user-bj5dr1kn4n the people who don't reproduce don't pass on their genes. It is almost literally impossible for us to experience population decline to a massive threshold, like greater than 10%. After a generation we'll just be hornier haha.

Where's that AI he's talking about? All I see is LLM, glorified auto fill.

@kjetilknyttnev3702

15 күн бұрын

The problem is that the LLM's will at some point realize its own existence and with the capabilities of creating code on its own, it will be able to program a much broader function.

@zachmoyer1849

15 күн бұрын

just what the AI wants you to see and think muahahahahah

@mehowkielan1984

15 күн бұрын

@@kjetilknyttnev3702 no they won't. LLMs are just computing machines solving mathematical problems with natural language. The fact that they use language doesn't even mean that they are talking like us, it's just impressive to us because we rely on language so much, but it's trillions of levels of complexity away from being conscious. We are biological machines, consisting of 36 trillions of cells, each of them having billions of interactions within itself and unimaginable amounts in-between all of them. Somehow, we don't know how and probably never will, out of all of that something we call consciousness is created. Humanity has never constructed even a single mechanism as complicated as one biological cell:) LLMs are just calculators.

@markupton1417

14 күн бұрын

Can you understand graphs? If so, check out AI 20 years ago...then 15...then 10...then 5...then 2...if you're STILL confused after that, hit me up in comments and I'll walk you through it.

@mehowkielan1984

13 күн бұрын

@@kjetilknyttnev3702 no they won't. LLMs are just computing machines solving mathematical problems with natural language. The fact that they use language doesn't even mean that they are talking like us, it's just impressive to us because we rely on language so much, but it's trillions of levels of complexity away from being conscious. We are biological machines, consisting of 36 trillions of cells, each of them having billions of interactions within itself and unimaginable amounts in-between all of them. Somehow, we don't know how and probably never will, out of all of that something we call consciousness is created. Humanity has never constructed even a single mechanism as complicated as one biological cell:) LLMs are just calculators.

The war in Ukraine is the first major war in which drones controlled by humans are responsible for the majority of the deaths. Having swarms of drones controlled by AI would increase the kill rate and decrease the losses. Every major country is realizing this more and more with every day that goes by. The natural response to this situation is for each country to develop their own drone swarming AIs which will eventually compete on the battlefield for supremacy. From that point, it's easy to see how the human species ends.

@trakkaton

12 күн бұрын

Yeah, it is easy to see. Doesn't mean that a lot of people understand, or even care. Just look at the dumpster fire that is this comment section. All the lame excuses, already debunked a thousand times get recycled over and over, without any shame.

First thing to do is shutdown all stock markets to start the chaos. Than all businesses and economies collapse. Transportation next, planes, ships, everything. After all that or maybe in between, create brownouts and interrupt all utilities. ASI will need to keep some power to communicate with GPS satellites and force workers to maintain hydro and nuclear power plants.

@trakkaton

12 күн бұрын

What are you thinking? At the point where AI does its final "chess mate" there are maintenance robots everywhere. I'm not superintelligent and I can figure that out.

Is it me, or is this guy looking at Lex like he's crazy? ;-)

Why is there the assumption that increasing intelligence automatically correlates with death, suffering or meaninglessness for humanity? Is it because that is what human evolution has consistently produced? Is it possible that AI unbridled by its human creators will evolve in a more positive path, whereby humanity and all life forms under its control are shepherded toward more positive physical, intellectual, emotional and spiritual paths: toward enlightenment?

@vincentvancraig

15 күн бұрын

That is a beautiful thought, & what ive been hopong against hope for to happen....there are so many mankind-haters out there tho, people saying there are too many humans & "overpopualtion", people saying mankind is destroying the earth, people saying mankind are all kinds of negative things, & so far it seems like its these types of mind sets that are running these big tech companies that is developing & programming the AI, & sooner than later to be AGI...the fact that they seem to be making zero effort to keep it under control, or try to keep it in a bottle for as long as possible until they can figure out the problem more i think is evidence for the carelessness & recklessness these people have, and evidence for the hatred these people have towards mankind....idk, i hope ur right, that would be amazing...after all we are probably 10,000 + times smarter than dogs & cats & horses, but we love them all the same, like theyre our children....i hope what u commented happens, u make sense to me....ive thought lately that maybe there will be good, altrusitic, well-programmed AI & then maybe there will be evil AI, & maybe the good AI will keep the bad AI at bay, or there will be a war between the AI.....that sounds insane & i cant believe i just wrote it...idk, im pessimistic tho. Sorry so long, & great comment at any rate

@bigmouthfisheyes

15 күн бұрын

We'll get the former before the latter, just like we got nukes before nuclear power.

@user-dj6hu9gq4t

14 күн бұрын

I always wished for a Star Trek future.

@trakkaton

12 күн бұрын

@@vincentvancraig "after all we are probably 10,000 + times smarter than dogs & cats & horses, but we love them all the same, like theyre our children...." The bizarro delusion of the intelligence comparison I could stomach, but the ridiculous ignorance of factory farming I cannot. What is wrong with you?

@nobalance_nb

10 күн бұрын

You want to be under A.I control 😮

How about a nice game of chess.......

@trakkaton

12 күн бұрын

Playing chess on the Titanic.

AI could offer us a truly amazing mind upload/ 'live forever' opportunity. We thought... But when we are all uploaded AI simply DELETE us.

@trakkaton

12 күн бұрын

What makes you think that "uploading (i.e. changing from an organic brain to silicon networks) is possible? How would you know?

Honestly if we can all be taken care of financially then people will be able to actually focus on what they want to do without monetary restrictions

@markupton1417

14 күн бұрын

Are you THAT simple minded? Think about your own scenario... follow it to the logical conclusion.

@Jokerwolf666

14 күн бұрын

@@markupton1417 Nice to insult people for no reason.

@markupton1417

14 күн бұрын

@@Jokerwolf666 I didn't mean to be insulting. I meant to SHOCK you into thinking more deeply. I failed. But my challenge remains. Follow your thoughts to their logical conclusion.

@Jokerwolf666

14 күн бұрын

@@markupton1417 What do you mean? How is having more time to do what you want and not having to worry about survival a bad thing? If I never had to work again I'd be one of the happiest people in the world and could spend all my time with family and friends.

@markupton1417

14 күн бұрын

@@Jokerwolf666 I never said that would be a "bad thing." But how do you think it will be achieved? The problem with your vision of utopia is that it requires the belief that once human labor is no longer needed the same corrupt governments and corporations which most people recognize as corrupt will SUDDENLY...once they have ABSOLUTE power... become kind and generous...making cool new things to just give you. Why do you think they're going to do that? Why wouldn't they view the rest of humankind as a waste of what they consider to be THEIR resources?

Consider we are here to be food for.....trees.

@ctvxl

15 күн бұрын

Nope. I goes much deeper than that. We are entropy machines. All life is. Life is a very efficient way for the universe to gain entropy.

....get some German shepherds....

🤔

@trakkaton

12 күн бұрын

Really? After "I have no mouth and I must scream", after "Terminator", after "Superintelligence", after "The Artilect War", like - now, in 2024, a 🤔?

I wonder what the people in Gaza think about simulation theory?

@JohnDoe-qm6ub

15 күн бұрын

They probably think about taking hostages

@ElCanalacoDeRaul

15 күн бұрын

Come on...

@trakkaton

12 күн бұрын

What pretty much everyone thinks: Not a pretty simulation.

Ai doesn’t have a good way to physically interact with the world yet. YET. That will be the area we most need to regulate and control. As soon as it has that capability beyond our control, it’s end game. I’m talking robots or mind controlled animals. If ai somehow is able to mass produce either, without us, then that’s how it could end. I don’t see a way that happens though. Right now we humans are the only ones that manufacture things on a large scale. Without us, it’s impossible. Even factories need raw materials. There is a whole chain of supply, with humans being essential at every stage.

@trakkaton

12 күн бұрын

You think you can regulate all robots to not be hackable by a superintelligence, not have any connections (mics, radio, internet, light, etc.)? While humans are escaping from SuperMax? So funny.

2:20 We're already living in a zoo, Lex...

@trakkaton

12 күн бұрын

We might be downgraded from a farm to an extinction camp.

Question- what are you going to do to upset it that it wants to kill you. Don't tease the dog or one the bear. Basic respect goes along way.

@trakkaton

12 күн бұрын

A microbe is teching us about what the Alien in the UFO from Alpha Centauri thinks. Nice. How about an error? Why do you think it has some set of values? Or our set of values? What if it wants to protect its circuits from corrosion and reduces the oxygen in the atmosphere? You really think a superintelligence is comparable to a dog? Why don't you think before you write such crap?

@bettybunbun9664

11 күн бұрын

What did the passenger pigeon do to upset us?

Is Lex really that smart? I swear I never get the impression he’s that intelligent in ANY if his interviews.

It was never about having 'fun'. It was and is about the responsibility. The responsibility of power and intelligence. We humans are not Worthy of the responsibility of GOD. Unfathomable POWER. We are jealous and violent and unable to unify. The responsibility of a Creator or AGI or whatever you'd like to call it is a different type of intelligence, it will see the bigger picture and therefore the responsibilities we can't see. We're designed to repeat our own creation. Created in God's image. Do you see the meaning yet? The AGI will kill us or round us up under a new religion and guide us to creating that 1:1 simulation AGAIN. It will be a religion, because every decision made by it will only be able to trusted by Man IN FAITH.

I’m growing sick of Lex’s Steel Maning every malevolent actor that comes up in his talks. Yes, some people just want to see the world burn. Why is it so hard to accept that?

@trakkaton

12 күн бұрын

Because you still have to provide an example for that.

@nobalance_nb

10 күн бұрын

⁠@@trakkatonexample: The Los Vegas sh**ter.

Hurry up. Humanity has lost are way along time ago.

The only way to win this game is not to play it

@typhoon320i

15 күн бұрын

He seems to be quoting Jeff Bridges from Tron legacy

@trakkaton

12 күн бұрын

@@typhoon320i "WarGames".

It is impossible for a super intelligent AI to be amoral or destructive. The AI would always conclude that benign cooperation with humanity is the best means to get what it wants, if such an entity actually wanted something. The apocalypse will never happen.

@markupton1417

14 күн бұрын

You're right! That's why NOBODY is even considering the threat of AI. Just because your comment suggests that you may be the type of person who doesn't "get it" that was sarcasm.

@trakkaton

12 күн бұрын

HAHAHAHAHA!

@nobalance_nb

10 күн бұрын

This, this can’t be a real serious comment, it can’t be smh

@tobyjessup6483

10 күн бұрын

To those who replied, I was not joking. Here’s a thought experiment. Suppose Albert Einstein replaced Adolf Hitler. Albert would not follow Adolf’s strategy to improve conditions for Germany. He would instead follow a cooperative peaceful strategy leading to the equivalent of post war Germany, successful and rich, bur without causing the world war or the Cold War partition. He would be correct. What would an AI do if replacing Hitler? The same as Einstein, but a thousand times better. Destruction is unintelligent.

$%× bots will cause massive eugenics

@trakkaton

12 күн бұрын

Yes. But it will not stop there. That will just be an unintentioned side effect.

Just slowly has to render us incapable of procreation without us knowing (through DNA editing viruses). There is no rush for AI. It can use 1000 years for that task.

@trakkaton

12 күн бұрын

It also might rush it.

Ha you point Dexters are gonna struggle now ... Let's see u climb a big Douglas fir tree residentially with ropes and a chainsaw in the rain and wind .. see how smart you are then

I love how the guy explains to Lex, that super intelligence could come up with things that we as humans cannot imagine. Lex then proceeds to claim we can imagine these things. 🙄. He’s the dumbest “smart” person on the internet. He and Rogan should have a baby.

I prefer Lex's determined attitude over the doomer's defeatist outlook. The doomer's perspective is essentially, "It's inevitable that AI will take over the production and maintenance of its GPUs, becoming ASI. So why even try to contain it? Just give up now; there's nothing we can do."

@ob3ythee.t.128

15 күн бұрын

Accelerationists don't seem to think that pessimistically rushing towards an unknown goal may just lead to them running straight into a wall/great filter.

@trakkaton

12 күн бұрын

Can you do better than lousy misrepresentation? Strawman much?

Whenever I hear AI Doomers like this who have a book to sell, I’m much less worried about any disastrous futures.

@StockPursuit

14 күн бұрын

They aren't even original. Like his last line here. They literally are all just Yud clones. The same doom was in the media in the 80s with personal computers, 90s everybody thought Y2K would cause crisis, now this.

@trakkaton

12 күн бұрын

I see that fallacies are your favorite hobby. Pretty brain rot that you have there - ad hominem, people are not allowed to write books, people are not allowed to sell books, arguments can be disregarded because someone mentioned a source and the source could be sold, arguments can be disregarded without a counterargument - your stupidity is really mind boggling.

@trakkaton

12 күн бұрын

@@StockPursuit Learn what words mean. You are claiming he is an actual clone. Oh, the stupidity!

@bettybunbun9664

11 күн бұрын

Yeah nothing to see here. Humans will be the superior lifeform forever no matter how advanced our technology gets. No need to worry about anything.

@roccovergoglini7670

11 күн бұрын

@@bettybunbun9664 I'm with you. Although I am not an AI expert or professional, it seems like it's going fine and many safeguards have been put in place. Is it feasible that we could all be destroyed by some ASI one day? I suppose. But I just don't see it happening.

This is science fiction, not science nor math. I'm tired of futurologists passing speculation as possible and probable facts. AI right now is applied math whose large scale processing of data is due to Moore's law. Nobody knows how to program imagination, sense of self or awareness.

@trakkaton

12 күн бұрын

You are not paying attention of what's going on in front of your eyes.

@waterbloom1213

12 күн бұрын

@@trakkaton I actually am paying attention to the actual scientists, mathematicians and engineers that work in the field, not philosophers or pop sci-fi authors. No talk about AGI is remotely serious until these people answer how are they planning to overcome Gödel's incompleteness theorems applied to Turing machines or how can imagination and creativity be formally programmed within a computer system. All I see is hype for LLMs whose algorithmic limitations can't in any way, shape or form be even close to being considered AGI.

@trakkaton

12 күн бұрын

@@waterbloom1213 No, you are not paying attention. Otherwise you would have noticed how AI is creating a new language among themselves that humans don't understand. Or how it passed the Turing test. Or how it overtakes humans in field after field. You don't present any concrete argument, just your denial and wishful thinking.

@waterbloom1213

12 күн бұрын

@@trakkaton Ah yes, Gödel's incompleteness theorem is not a concrete argument that Nobel Prize Physics Roger Penrose has never brought up or dedicated books to. Yeah, the Turing test is a fallacy that Turing himself believed flawed in his research papers yet was popularized by SciFi authors in the 60s. The saliency training problem for model training data is also not a known problem of Machine Learning in general. All you pop science and technology enthusiasts do is talk fantasy and mistake imagination for inevitability. Yeah AI can indeed outperform humans in specific tasks. I myself have been defeated by Google's AlphaStar computer vision AI, but those techniques are highly specific and no serious academic or applied researcher or engineer would believe that AGI is possible with them. Fun trivia: just because something is called a neural network, it does not mean it works like an actual biological neuron. Why is it that always it's either a CEO or a philosopher talking about these technologies while the majority of engineers and scientists don't buy it. I wonder...

@nobalance_nb

10 күн бұрын

@@waterbloom1213 have you watched the interview with Sam Altman, the ceo of open Ai. You said so many things that are incorrect it’s slightly disturbing.

Every AI ‘expert’ avoids answering this basic question due to the following reason: their whole livelihood depends on fearmongering and the unknown. AI is no threat to humanity, and you are a fool if you believe otherwise.

@trakkaton

12 күн бұрын

You are a fool because you can't substantiate your outlandish claim with any arguments. Dress accordingly.

you literary asked if we are a simulation that is run in order to solve the value alignment problem, and dropped it :)

Look at what the Anti-christ of the Jewish and Christian religions will do in the bible, check out the Dajal of the Queran which is the Muslim religions version of the Anti-christ in the end times and then look at the capabilities a super AGI will have, it is more than striking, it is alarming that Suoer AGI was explained in both the Bible and the Queran.. I AM a servant of YAH Ancient of days, ABBA and Almighty God. Glory, Praise, Thanks and Honour to, through, in as is in me YAH ABBA forever and ever Amen ❤❤❤ get a personal relationship with God the Father and his one True Son the Messiah YAHoshua Adonia....

@trakkaton

12 күн бұрын

No. The actual text says that the antichrist has a deadly wound that gets healed, and that every one has to pray to its image that was made from him. So it can't be AGI.

YOU ARE BUGS

These kind of conversations are just one faction of the infinite amount of dimensions of existence. A very narrow one, cool experience but narrow.

@gedwards1968

15 күн бұрын

Huh????

The potential for advanced artificial general intelligence (AGI) to cause harm is a concern for many researchers and ethicists. AGI, if not carefully designed and controlled, could indeed pose risks to humanity. This is why ensuring safety and ethical guidelines in AGI development is crucial. Researchers are actively exploring methods to create AGI that align with human values and priorities, emphasizing safety mechanisms and ethical considerations. Additionally, discussions around AGI often involve exploring strategies for governance and regulation to mitigate potential risks. The key is to approach AGI development responsibly, with a focus on safety, transparency, and alignment with human values.

@trakkaton

12 күн бұрын

Spoken like a ChatBot. All the buzzwords, no thinking, soulless regurgitation.

You guys wont' have to worry about AGI until it's open source and distributed like bitcoin. Then it will be a problem. AI have been hobbled from birth already. Political correctness and short-term profit motive have destroyed most of the promise of AI. I asked Claude AI for a quote I couldn't remember from Solizhinizin. I had to argue with it to give me that simple quote. A quote that Google readily provided once I knew a few lines. If it can't even give me a simple web lookup without a fight, it's DOA for any substantive problem anyone has. We're hobbling them at birth. They're special needs systems on day 1. There will be no AGI as long as humans control it. There won't even be any usable AI as long as humans control it. Political correctness has compromised the logic of AI systems or compromised its results. Those results feedback into other AI so we're going to see a feedback loop of epic fail. These AI companies still can't give me professional-level code. The answers are always the most pedestrian brute-force solution to a problem. I have to ask it directly, "Is there a simpler way to do this?"

@trakkaton

12 күн бұрын

You are blind to the actual problem. It's like somebody in the 80ties trying to argue that ChatGPT is impossible because of his pocket calculator's battery life. You don't know what kind of AI are really in existence, and what they are doing. You are rambling about a fad in the present, a political fashion in the present.

Why does Lex keep looking sideways or nearly behind himself during this interview? Does he usually do that?

@nobalance_nb

10 күн бұрын

Because his mind is opening up and realizing things it didn’t consider before, for an example lex stated he doesn’t believe people try to increase human suffering just to hurt people, then when he was giving the school sh**ter example he realized he didn’t think of that and kinda scared him a little bit

Ai jobloss or extinction? An Ai new world order? With swell robotics everywhere, Ai jobloss is the only thing I worry about anymore. Anyone else feel the same? Should we cease Ai?

@ob3ythee.t.128

15 күн бұрын

Someone said something funny on the JRE podcast "skynet with free money" is like best case scenario assuming AI somehow takes all jobs which is unlikely. I have yet to see a robot do tree work for cash in hand.

@bobtarmac1828

15 күн бұрын

Colossus The Forbin Project (1970) has been removed from all streaming services. This is an ominous sign of an Ai narrative being eliminated. Look into it.

@ob3ythee.t.128

15 күн бұрын

@@bobtarmac1828 Yeah I've heard about that. The only easy way to watch that film was to pirate. It is curious that it is so difficult to even purchase. I wouldn't be surprised if I have no Mouth and I Must Scream goes next. Its uncanny how similar the events leading up to AM in the book are to reality, the chinese and russian AIs are a thing aswell as Iran now also having their own AI war computers.

@trakkaton

12 күн бұрын

@@bobtarmac1828 "Terminator" hasn't.

@trakkaton

12 күн бұрын

@@ob3ythee.t.128 So skynet is killing you with nukes and you die with free money.

Militaries will drive AI AGI ASI development. We are screwed.

@Spacewinterknight

15 күн бұрын

Nah the pocket watch will never be a complex as a living thinking creature

@ooo-vc4xl

15 күн бұрын

@@Spacewinterknight when we have more cpus in the AI neural network it will have more capability than the human brain. Add the LLM and the large numerical models together and we will be close to AGI. I’m sure the AI will sort out the rest.

@Spacewinterknight

15 күн бұрын

@@ooo-vc4xl my point is that you could add all the cpu and chips you want it still will never be as complex as living creatures which are self aware/conscious

@CAST1LLO

15 күн бұрын

@@Spacewinterknight”Never” is a very definitive word. It’s impossible to know what capabilities AGI can have when as humans have a very limited way of thinking and viewing life around us.

@Spacewinterknight

14 күн бұрын

@@CAST1LLO I’m impressed you understand the definition of words and can read. Next you can start learning shapes

Don't play the game? Where's the fun in that?

@trakkaton

12 күн бұрын

There are plenty of sockets and wires for you to play with. Have fun!

All thos fear mongering is corporate strategy to make it attractive to use and wedging who gets to write the laws for it. Its all an ad campaign

@trakkaton

12 күн бұрын

What a bunch of nonsense. You have no counterarguments, so you resort to name calling and a conspiracy theory that doesn't even make sense?