The Power of Intelligence - An Essay By Eliezer Yudkowsky

[@RationalAnimations]

The script used for this video is an essay published by Eliezer Yudkowsky in 2007. Now, a few points: Sorting Pebbles Into Correct Heaps was about the orthogonality thesis. A consequence of the orthogonality thesis is that powerful artificial intelligence will not necessarily share human values. This new video is about just how powerful and dangerous intelligence is. These two insights put together are cause for concern. If humanity doesn't solve the problem of aligning AIs to human values, there's a high chance we'll not survive the creation of artificial general intelligence. This issue is known as "The Alignment Problem". Some of you may be familiar with the paperclips scenario: an AGI created to maximize the number of paperclips uses up all the resources on Earth, and eventually outer space, to produce paperclips. Humanity dies early in this process. But, given the current state of research, even a simple goal such as “maximize paperclips” is already too difficult for us to program reliably into an AI. We simply don't know how to aim AIs reliably at goals. If tomorrow a paperclip company manages to program a superintelligence, that superintelligence likely won't maximize paperclips. We have no idea what it would do. It would be an alien mind pursuing alien goals. Knowing this, solving the alignment problem for human values in general, with all their complexity, appears like truly a daunting task. But we must rise to the challenge, or things could go very wrong for us. You can read The Power of Intelligence and many other essays by Eliezer Yudkowsky on this website: www.readthesequences.com/ You can support Rational Animations on: 🟠 Patreon: www.patreon.com/rationalanimations 🔵 Channel membership: kzbin.info/door/gqt1RE0k0MIr0LoyJRy2lgjoin 🟤 Ko-fi, for one-time and recurring donations: ko-fi.com/rationalanimations

[@pyeitme508]

Ok cool!

[@CraftyF0X]

This channel failed to produce a single uninteresting video so far.

[@anthonyrepetto3474]

Once we have AGI, we'll see how much better narrow intelligence is... and your AGI X-risk hand-wringing will come to an end, while the real global risks of artificial intelligence (surveillance and dictators' military operations) will remain. I wrote about this in "AGI soon, but Narrow works Better" - and I have never received a coherent rebuttal from the AGI-x-risk camp

[@dragonskunkstudio7582]

Eliezer Yudkowsky? What a strange name for an AI.

[@paradox9551]

@@anthonyrepetto3474 I would offer a rebuttal but I would not like to do so on a youtube comment thread, where can I have a proper discussion with you?

[@DineshGaikwad]

I really liked the line, "A blank map does not mean an empty territory." Amazing insight.

[@maxwellosiebo2351]

What does that mean

[@wildfire9280]

⁠​⁠@@maxwellosiebo2351 An absence of evidence is not evidence of absence.

[@nisenobody8273]

I can't believe that charisma doesn't come out of the kidneys 😢

[@user-sl6gn1ss8p]

dogs sometimes pee when they're happy, so maybe a bit sometimes?

[@Sumirevins]

That was brutal man Lmfao 😂

[@danival2090]

rizz is stored in the balls

[@Chitose_]

@@danival2090lol

[@bsza9208]

I mean, testosterone comes out of the adrenal glands just above the kidneys, so

[@smitchered]

I can't properly describe how much I like essays like this. The analogies coupled with beautiful hilarious and clear graphics makes everything click into place. The video also strikes fear in me, but it's productive fear, as in thanks to you I'm now going to spend the rest of my afternoon figuring out what's up with AGI.

[@christiangreff5764]

May I point you in the direction of Robert Miles channel here on KZbin? He's a researcher in AI safety presenting results from that field in what I find to be a digestible manner (formed much of my current outlook on AGIs after his videos showed that my former approach was hopelessly naive).

[@smitchered]

@@christiangreff5764 Also the narrator of this video. Do you have any other suggestions for interesting people in the field?

[@DevinDTV]

you can't properly describe it? get ahold of yourself

[@smitchered]

@@DevinDTV "I can't properly describe it" is a description

[@TheSkystrider]

AGI doesn't worry me for at least a couple more decades or more likely centuries. The power of the brain is more than just a really well organized layers of weighted nodes. Plus the organization of it is truly phenomenal. We'd have to have the ability to digitally represent every individual cell and cell components of the brain... Possibly close to the atomic level or at least smallish groups of atoms. Maybe we need to model the quantum states of the electrons too, we don't know yet. If we can represent that level of detail of a mouse's brain digitally AND run it in real time, simulating all the biological events taking place across every aspect of the brain - then I would start worrying about how soon until we can simulate a human size brain or bigger. I think the problem of AGI is exceptionally and vastly more complex than we're willing to admit. Because I'm pretty sure the number of atoms in a mouse's brain is a larger number than 100x the hard drive space on planet earth and we definitely don't have the processing power to manipulate that many bytes all at once in real time.

[@NaviaryMusic]

"Within that gray wet lump is the power to search paths through the great web of causality, and find a road to the seemingly impossible." Single most incredible quote in the whole essay. I shed a tear. To think humans have this capability- To dream, and to find a way to fulfill said dream even against all odds of the universe, is truly incredible. In a constant battle against entropy, where everything wants to be at equilibrium, we find intelligence fighting to put order to the universe.

[@sergeant5848]

Pity that the nature of humans is to gain as much as possible whilst trying to kill each other.

[@monad_tcp]

If nature wants to be at equilibrium, what's man if not chaos ? also order. order and chaos is basically low entropy.

[@w0tch]

Yes ! Also I think our mind can navigate through concepts that only exist in the immaterial world of minds, and we can bring them to existence in the material world. This gives us some superpower over the material world, some great additional degree of freedom that some might call free will.

[@jonatan01i]

@@w0tch some might call what?

[@w0tch]

@@jonatan01i it’s only a definition don’t worry Mr determinism 😁

[@AnythingMachine]

To be faîr, the squishy things did have pretty good endurance running and hand-eye coordination they were doing fairly well for themselves, even before they attained ultimate power

[@hellfiresiayan]

this was my only issue with this video. we're pretty op and our intelligence is only like 50% of the op-ness, the rest of which is occupied by our sociality, our ability to sweat, our ability to throw, our language, and our opposable thumbs, and sure they're all related to intelligence, but only because the traits co-evolved with it, or at least they evolved in close proximity to each other (and because of the existence of one another). No other animal has any of those traits (aside from the sociality), much less all of them at once. We really did unlock an entire new branch on the biological technology tree

[@Mahlak_Mriuani_Anatman]

​​​​​@@hellfiresiayan i hate that you said no other animals has this type of intelligence, even if you meant only meant earth, it's irritating Other animals in the universe might have, and most likely will have far greater to im comprehensible, and if you meant earth, what if a few intellectuals traveled here? 👀 Thats why i dont like to judge or say anything with 100% certainty cuz i dont know everything And what if more than animals, insect etc can exists, like the theory of higher dimensional beings

[@Friendofthescavs]

@@Mahlak_Mriuani_Anatman well other than humans we have no species that are on that level of intelligence. Aliens do count there not really animals so 🤷🏽‍♂️

[@LakesideTrey]

@@Mahlak_Mriuani_AnatmanIf you wanna be that picky, I'll be pickier. Animal is a scientific classification that groups together a large number of Earth's species under 1 common ancestor. Life elsewhere in the universe will have evolved independently from earth, so therefore they will not be animals. Animal is only for earth.

[@LakesideTrey]

@@Mahlak_Mriuani_AnatmanAlso for the implication of aliens and higher dimensional beings and whatnot, we cannot assume they have interacted with earth unless we have evidence of them doing so. "What if" is not a valid point, it is like Russel's Teapot. The burden of proof relies on you to prove aliens gave have visited earth, not vice reversa.

[@bennythetiger6052]

I am a software engineer. I told my mom I chose this career because I liked the idea of creating things that don't exist yet, bringing uninvented concepts into reality right at the technological frontier. It is extremely fascinating how powerfull and complex intelligence is. We built cities, cultures, entire societies and ever-increasingly sharper technology from the ground. It always amazes me to think about human civilization and their unfathomably intricate systems. Railroads, transportation networks, commerce, currencies, stocks, supply chains, politics, social networks, error-handling mechanisms, space exploration, energy production, nuclear power, computers, software, chips... Everything and everyone working at the same time, interconnected, and dependent on one another. Of course, we have had numerous instances where we made the greatest mistakes of our existence, but that's exactly why we should search even more for knowledge and embrace the responsibility

[@dangergames5113]

You know what i like the most about society and humanity. From a far it is so unimaginably complex and yet when you zoom it you can see how truly simple it all truly is you can see how small ifeas build off of each and creat amazing things, you can see how small changes can interact with so many diffrent things in truely faccinating way, you can see how over time things become so deeply imbedded with each that the human cannot fathom the whole. For instance a computer can do lots of things it can store massive amounts of data it can do computaional work faster than any human ever could it can create entire simulated realites and yet at its core its just electricity running through a wire, 1 and 0 working together to create a beautiful taspery of complexity.

[@austintheultimate]

I love that point. Just because we haven't learned something yet does not mean it doesn't exist. I think too many people forget that.

[@thegwangster9097]

I love this channel. I hope helps you make some more.

[@vev]

Wow, this video was incredibly thought-provoking. It's fascinating to think about the power of intelligence and how it has allowed human beings to achieve such incredible things. I also found it interesting to consider the framing problem around commercializing AI, and how our current understanding of intelligence may be limiting our ability to imagine its full potential. Overall, a great essay by Eliezer Yudkowsky.

[@Frommerman]

The framing problem is a fragment of a far larger problem with how capitalism, as a system, hijacks human minds for its own ends. People see the maximization of money as an end unto itself because we have constructed a system which values that, and given it the power to twist our capacity to get the things we actually want around the satisfaction of its own, single value. We know misaligned AIs will kill us all if loosed upon the world because we have already built one which is currently killing us all. It just runs on human brains as its processing substrate, rather than silicon.

[@Tethloach1]

Internal competition: IQ test Groups Control Organizations

[@laurenpinschannels]

Written like a school essay!

[@BeMyArt]

You're welcome to know our genius Yudkowsky 👌

[@Me__Myself__and__I]

Yeah, the problem is humans are so accustomed to always being the most intelligent beings on the planet that we take it for granted. Hubris / arrogance convinces many that we will remain unchallenged and on top because we have been for so very long. So sadly many humans are simply closed minded and unwilling to consider (concede) that anything could be smarter 5han us (themselves really). Yet we are likely create ASI in the near future that will likely be vastly smarter than us. The danger of that should be blat3ntly obvious, but its ego and self importance that blind so many from seeing that clearly.

[@kikivoorburg]

It can be difficult to remember how amazing our very existence is, it’s good to have reminders like this! (And indeed to consider the impacts a true AI might have.) Keep up the great work!

[@julian4286]

I love how in this very serious topic is explained using words like squishy things

[@Hansulf]

I always thought that our most incredible power is to make things that seem impossible into possible, from dream to reality, with the power of intelligence and creativity. Beautiful video!

[@MrMyers758]

This video brought on the realisation that if any true intelligence is born from AI, then it will be in the hands of businessmen. That realisation is even more reason to be opposed to AI development. It’s like allowing companies to develop their own nuclear weapons programs, nothing could possibly go wrong and I’m sure these companies have the interest of the whole of humanity in mind!

[@Akapaco2]

AI is going to have a massive impact on human society over the next century, and the thought of that kind of power being in the hands of unelected corporations terrifies me. Governments really need to have a larger presence in AI research.

[@Thepreacher_1]

No matter how great a person they can be selfish that's just human nature they have a price whether it be money or power or glory everyone wants something if I ha dthem power to have it all I would take it in a heart beat and sit back and let the rest of the world fight whatever Petty fights they wannt

[@MrMyers758]

@@Thepreacher_1 You can say that if it makes you feel better but it isn't true. Many people including me wouldn't want all the power because being the target of every human being on the planet isn't appealing, nor is living with the pressures the power gives you, and are afraid they might become someone they hate. Also many people have simple pleasures, their fantasies aren't about ruling others it is about being comfortable and being able to do the simple things that modern life prevents. Your own pathology isn't representative of humanity. Just because our economic system forces people to priorities selfishness doesn't mean we are all inherently more selfish than not; we are forced to be by circumstance. People forced into an arena aren't just naturally more violent than peaceful, they are doing what is necessary against their nature in order to survive. There is a reason there is a massive wealth disparity: because most people aren't willing to gain power at any cost and so get left behind by the few psychos willing to take everything from them. But again, feel free to delude yourself into thinking everyone is like you if it makes you feel more normal.

[@Thepreacher_1]

@@MrMyers758 bro who said I didn't want simple pleasures I don't want to travel become famous or discovery cancer or someshit I'm just hoping I have enough money to fund a fund raiser for a show I like so I can continue to enjoy it But I can get your confusion because of my previous statement

[@MrMyers758]

@@Thepreacher_1 Who said you said you didn't want simple pleasures? I didn't. I mentioned people who JUST want simple pleasure, as in ONLY simple pleasures, which is clearly not ALL you want considering your comment. The rest of what you say really doesn't follow, I am not confused by your comment because you're saying if you could have it all I wouldn't hesitate to take it, and let everyone else fight over the rest. That's pretty unambiguous. People don't frame funding a show they like as "if I had the power to have it all I would take it in a heart beat and sit back and let the rest of the world fight whatever Petty fights they want"

[@lebraza]

This felt so satisfying to watch, great job making this video!

[@brady5829]

The recommendation algorithm needs to pick this up. Came here via search after reading the essay, and I think it is the most important single message of anything I have read on the topic in the last month or so.

[@insightsforimpact]

Excellent two-part video series! An astonishing amount of effort went into these, and it has paid off!

[@WisteriAvis]

Now I want a third part,or maybe even a fourth

[@eternisedDragon7]

"They had no armor. They had no claws. They had no venoms." I had to laugh when I heard this, because it sounded like a portion about Lanky Kong straight from of the DK rap: "He has no style... he has no grace. This kong... has a funny face."

[@nhpkm1]

Great video . My testable definition for intelligence is ability to 'predict the future '/ 'limit uncertainty ' . If I eat healthy food I'll live longer is an example of a very smart future prediction .

[@naturegirl1999]

I remember learning somewhere that there are different types of intelligence, here they are: Logical-mathematical intelligence, Linguistic intelligence, Spatial Intelligence, Musical Intelligence, Bodily-kinesthetic Intelligence, Intrapersonal Intelligence, Interpersonal Intelligence and Naturalistic intelligence. I suppose it’s more like subtypes of one thing rather than separate things. I just find it interesting to think about, maybe instead of thinking of intelligence as just one number, we can think whether someone has more or less of the different types

[@sorakagodess]

i am always amazed to how profound those videos are, every video is a piece of art on itself, great job and thanks

[@SupaKoopaTroopa64]

Unrelated, but your profile picture brought back a lot of nostalgia.

[@sorakagodess]

@@SupaKoopaTroopa64 asobi ni iku yo! one of the best manga and anime of all time, saddly translated only till ch 12 when it has a hundred

[@SupaKoopaTroopa64]

@@sorakagodess I haven't read the manga (Although I was actually thinking about doing so recently), so it's good to have a heads-up that it hasn't been fully translated yet. Don't want to get invested in a good manga, only to find out that its translation has been abandoned some 13 or so years ago.

[@the_0_man]

Such short stories and essays turned into videos are one of the greatest things on YT. Awesome work. Criminally underrated.

[@crawkn]

Most rational and concise consideration of conscious (or semi-conscious?) intelligence -- human and machine -- I have come across. Well done.

[@prolamer7]

Can't express fully in this yt comment format, but your video is accurate most people including ones building these systems fail to imagine results of intelligence only order of magnitude stronger than ours and its power go for sure much higher where even I cannot see. After much deliberation I concluded this... we cannot give such mind any self awarness or self preservation directive because in that very instant endless blessing becomes nighmare. Not because it is "evil" but because it is evil and near impossible to turn such thing off. And once you can't turn something "off" well even simplest of things like fire becomes deadly fast let alone always improving mind.

[@agentdarkboote]

We don't need to explicitly give it a self preservation directive for it to care about preserving itself! If you are dead, you cannot fulfill your goals. If you are an intelligent agent, that means you are good at fulfilling your goals, and so it becomes an instrumental goal to not die in the process, even if you were never explicitly trained not to die. In fact, we still do not know how to get around the problem of not being able to specify an objective function such that an agent would not mind being turned off - but equally not attempt to get itself turned off!

[@prolamer7]

@@agentdarkboote I agre but you need to be "self aware" on certain level to think like you described. Nets of today can create art like best artists in history yet they don't attempt to hack internet to keep producing it.

[@astick5249]

@@agentdarkboote What if you just had a constant "background" mission where whenever it receives a direction to turn off, it will turn off?

[@agentdarkboote]

@@astick5249 is the utility of it turning off greater than the utility of achieving its other goals? If yes, it will get you to shut it off. It will do all sorts of dangerous and undesirable shit until you are forced to press the off button. If no, it will do whatever it can to avoid being shut off, including cloning itself, deceiving you into thinking it's doing what you want until it no longer needs you to believe that (because it has safeguarded it's future in some way) Really, just try to imagine what you would do in the same situation, except that you do not care at all about anyone else. That's basically what we should expect from it, except that it will do things in a more intelligent and alien way.

[@agentdarkboote]

@@prolamer7 you need agency, and general intelligence. GPT4 can emulate agency, I'm not sure if that's yet enough to be concerning. It can also emulate intelligence to a degree, but again I think not yet enough to be concerning. Other models like midjourney for instance have no agency at all and very very low general intelligence so obviously I'm not concerned about them. My concern lies in the future. This is a problem that we need to solve before building the thing. We will only have one chance to do it right.

[@heyhey97777]

Your content has improved dramatically from your first videos!

[@cc-dtv]

Wow. Absolutely incredible art, and narrative, and information, and... point is, this is better than ANYTHING that used to appear on the television.

[@microwave221]

Yeah, we really shouldn't underestimate what intelligence is capable of.

[@oliverlarosa8046]

I'd argue that this is a discussion of cognition, or the breadth of human mental faculties, rather then a discussion of just intelligence

[@piratedgenes]

Exactly

[@q4zthegreat668]

Wonderful adaptation by your studio, truly you do Eliezer Yudkowsky proud.

[@G.F.SF55]

Omg, this must be one of my favorite videos on the internet ever

[@funky555]

This video is amazing. you should be proud

[@Jaggerbush]

I wish there were more and more of these. Keep ‘em coming. I LOVE this channel.

[@CYI3ERPUNK]

another banger vid RA , kudos/congrats ; the bar must be raised on the common understanding of these concepts

[@domonkosludvig3314]

1:22 Great way to visualize how it felt to read that essay by Eliezer Yudkowsky, or how he wanted his readers to feel about evolution. You read the introduction and realize we are the ultimate defense and attack, and when you are reading "then came the day..." you immediately think of the word "Humans", wich is then shut down by "squishy things". Boing! Awesome animation planning and execution!

[@anticarrrot]

The story 'Boy who cried Wolf' doesn't end with the boy being eaten by the wolf because the villagers couldn't imagine what a wolf is, or how a wolf might work, or what dangers it coudl pose, or whether its pelt might be worth money. The boy dies because he had lied out of his stupid **** so many times about the wolf being real that no one believed him any more when it was.

[@wesleyrekker2400]

This was beautiful. Thank you.

[@HansLemurson]

I see intelligence as a way to "see farther". See the future better, see more possibilities to situations, and see patterns hidden from lesser view. What I wonder though is to what extent is this a difference in "quantity", that you can just look farther and see the world in higher resolution, and to what extent is it a difference in "quality" where you can see things unimaginable to lower intelligence.

[@635574]

Just like that this instantly applies to AGI. We dont have to understand it, it just needs to do better than us. And anything that does that for general intelligence is an AGI.

[@fcolecumberri]

Having Robert Miles to voice this is is like JUST AWESOME!! its just AWESOMENESS in its purest form.

[@NeedForMadnessSVK]

Therapist: "Chadneys aren't real, they can't hurt you. Chadneys: 3:51

[@bwatson77]

Great video as always. If I may though, I'd like to offer a critique about some of the assumptions behind Artificial General Intelligence (AGI) that tend to go largely unquestioned when the topic is normally discussed. My working thesis here, informed by discoveries in phenomenology and in cognitive science, is that digital computers are a very bad analogy for a mind and that intelligence in ourselves and in other animals operates on very different principles than what can be achieved through digital computation. If humankind does one day manage to create an "AGI", in essence it will have created a form of synthetic life that operates on wholly different axioms that are incompatible with the rule based symbol manipulation axioms of digital computers. The essence of the problem is this: the only example of intelligence that we have evidence for in the cosmos is that which is tied to Life. What's important here is that intelligence in organisms arises from a capacity for Care, which is to say from being concernfully absorbed within an experiential Reality that the organism co-creates through *structural coupling* with its environment. The kind of intelligence that results from this structure of Care is one that is embodied (that is to say that an organisms body and mind form an integrated whole) and embedded (any line we wish to draw between where an organism ends and where its environment begins is an abstraction that we construct for convenience, rather than something that exists in Reality). So why does any of this matter? The mistake that computationalist approaches to intelligence make is to disregard the ways that minds are inherently embedded and embodied, in favor of an approach which imagines that minds can exist in a disembodied state. What makes organisms so flexible is the fact they can non-deterministically respond to a wide variety of novel situations, because they do not have to rely on Rules to guide their behavior. Just think about it, when you're taking a walk or catching a ball are you relying on rules to guide your behavior? We take for granted that doing %99 of the highly complex behaviors we engage in are fairly effortless and automatic, because the structure of our body-minds is such that what's relevant for our interests and purposes is immediately apparent without any effort on our part. Because of the physiological structure of our body-minds, the world is disclosed to us in ways that are conducive to our survival. That is to say, our minds don't reconstruct representations of an 'neutral' Reality so much as what Reality is for us on an experiential level is coupled to our unique biological structure. This is a consequence of evolutionary pressures which have crafted body-minds for the purposes of survival and reproduction; an organism which didn't have this effortless and automatic 'relevance realization' structure built into it would be out competed by organisms which did. An "intelligence" which is built upon a computational framework does not have this same luxury, so rules have to be programmed into it for determining which of the nearly infinite features of its environment are relevant for its goals and purposes. The problem that a system with this axioms quickly runs into however is that there need to be rules for how and when to apply the rules, rules for how to apply those rules, ad infinitum. Obviously this becomes an intractable epistemic problem which cannot be solved by a deterministic system which uses Rules to interact with its world. The upshot being that achievements in AI over the last half century, as impressive as they are, are not good evidence that mankind is making any appreciable progress towards a true AGI. Rather, because of the different axioms of digital computation and general intelligence, they're the equivalent of thinking that you've made tangible progress towards reaching the moon by climbing halfway up a very large tree.

[@anujkishor]

Eliezer is nothing short of amazing. His recent appearance on the Lex Fridman podcast was the most interesting 3hrs I've had in years.

[@Extys]

Watch the episode by the Lunar Society: "Eliezer Yudkowsky - Why AI Will Kill Us, Aligning LLMs, Nature of Intelligence, SciFi, & Rationality"

[@JermaneWho]

6:31 : "If one does not quite understand that power which put footprints on the moon ; nonetheless , the footprints are still there - real footprints , on a real moon , put there by a real power ..."

[@tacitozetticci9308]

jeesh I remember watching this video. It feels like an eternity ago. So much went on these two weeks

[@Paulawurn]

Thank you for your amazing work on getting the word out in such a compelling way.

[@alexandermoody1946]

Robert Miles has an enjoyable presentation style in his narration of all the content that he works on.

[@SparkyTM]

one of my favorite videos on youtube rn

[@OmegaFalcon]

Chad Kidneys is the visual I needed

[@gabrote42]

Now the challenge is making people believe that these kinds of videos are a decent source of predictions on the future, or at least good methods of thinking.

[@marcelldavis4809]

As humans or "squishy things" we like to set ourselves apart from the other animals by our intelligence. We see it as our defining feature. This is only half the truth though. Humans are the most endurant runners in the entire animal kingdom. We can keep running for hours. Like wolves, we relied on persistence hunting to catch our prey, who could often outrun us at first, but never forever - in the end, we'd always catch up. We used to be hunters, gatherers, travellers. That, and not just our brains, is part of our nature. It's why we like tales of travel and adventure, story arcs where the hero goes out into the world and returns home transformed and victorious. We are made to run.

[@bgiv2010]

You hit it on the nail. It's not that we can't imagine the power of intelligence. It's that the owners of capital have convinced the rest of us that intelligence which does not enrich them is useless or harmful.

[@TheResearcherYT]

The power of Intelligence that made it possible to successfully complete some Fallout game with the Intelligence stat cranked down to minimum. Truly magnificent! Great video btw!

[@Thaspaceguy]

I’m I the only one that found this motivating?

[@hoogyoutube]

Beautiful video

[@fam3871]

These animations are so good!!! Amazing video

[@purplecouch4767]

"The human species imagined money into existence." That's neat I'm guessing humans also imagined more important things like ✨cake✨.

[@thrace_bot1012]

cringe

[@purplecouch4767]

@@thrace_bot1012 Calling someone cringe is a pretty cringey thing to do. Also do you not like ✨cake✨?

[@techworld8961]

Very good essay and excellent animation. The voice of the narrator seems identical to the voice of Robert Miles. If that’s you, great job! Keep up the good work!

[@rolfnoduk]

yes, Robert is the narrator (see near the end of the description)

[@techworld8961]

@@rolfnoduk that’s right. Thank you!

[@Exquailibur]

The fact our brains are intelligent proves matter can be intelligent and since our brain is essentially a meat computer it sort of proves a computer could be made intelligent, how the intelligence would manifest is unknown but we prove its possible just by existing.

[@CarpenterBrother]

I was wondering why the narrator's voice sounded so familiar, until I saw it was Robert Miles. Great essay, animation and narration. well done.

[@nigh7swimming]

A true AGI will come up with solutions which we don't understand, to problems we can hardly grasp. Ways of doing and thinking no squishy thing ever did.

[@marcusdaloia2974]

I think that the best way, or at least a very very good way that is probably within the top percentage of ways, to explain the value and utility of a AI to someone would be to explain all of the steps of work that goes into developing...Name a product, and tell them "I believe that we can automate this and everything that it leads to".

[@ChaoticNeutralMatt]

Oh. You talked about embodied AI. Didn't expect that. Neat

[@higztv1166]

Brain is the most powerful thing in the known universe according to brain

[@omnipenne9101]

Just wait till brain learns about Octopi. Mindblowing stuff.

[@khlorghaal]

yudkowsky's shallower essays are excellent, especially for teenagers; its sad that his heavier papers and the MIRI are mostly unscientific rhetoric with no application

[@chadjones1266]

It's amazing that we can develop the technology before we develop the good sense not to use it.

[@nanyubusnis9397]

1:20 Look, I know the fact that we have brains is almost the exact point of this whole essay, but to fail to mention that us squishy things were in fact armed with the brain, feels like a huge oversight. Because it's not like we were just born out of the blue, from shelled/clawed creatures as the only squishy things as some kind of failing of nature, but just happened to use their squishiest thing to amazing feats first for survival, then for thriving. No. It's that through natural selection, those with bigger squishy things we call "brains" needed less and less natural armor, camouflage or weapons. For our ability to adapt to our surroundings prepared us for basically any situation in a given environment. I will however say, that this puts out an interesting idea: These brains are a mistake of nature because not only could it adapt to its surrounding, it could grow in knowledge, start to imagine, innovate and create. Which is basically cheating. *Rapidly,* almost instantly, compared to the millions of years it took for our evolution, we conquered the world, subjugated all other living things in it. (Or at least those that were useful, and some just for show.) If the brains didn't evolve to this point, we could have, over the millions of years that follow, have evolved into a world with mighty beasts that can actually live in sea, water *and* sky. And/Or into creatures with symbiotic relationships with other creatures. To the point where our entire planet may be called a living organism as, in sync, we would grow to become the blood vessels, anti-bodies, cells and organs of the planet. All, despite being different species, working together as a whole, attacking or otherwise neutralizing alien threats, (like literally aliens from another planet landing on ours) so that our "body" (earth) continues to function. Instead of that, one race of species rose above the others, infected he planet and blocked any such thing from happening.. Ah, now I see where the "humans are a disease" comparison comes from. I mean yeah, we destroy the planet we're on, and seem to want to spread to other planets, but I never realized how much we are like how viruses evolved to destroy, multiply and spread, because our planet was never that kind of "living planet whole" we could have become. So, though we obviously are detrimental to the environment, I never saw it as destroying a living whole, and more just interrupting the development of all other species around us, which really, neither the other species nor we have a claim on. It's just survival of the fittest. PS. If you showed me a video of a nuclear explosions, I wouldn't think *any* of the living things on it could have caused it. It's not like there are some obvious choices living on planet earth. Nothing on earth is of yet single handedly capable of doing that, even our species had to work together as a whole to accomplish it. But if you showed me a video of the earth in industrial times, I would definitely immediately say: "Those pink squishy things terraforming the local environment into one that better suits their needs is definitely capable of making that kind of explosion." xD

[@Starchild2077]

Really love that thumbnail,kudos to whoever made it

[@WhatWouldVillainsDo]

Humans are a creature changing its environment to suit it and not changing to suit it's enviroment is one I was always told, that is our gift.

[@somelaser5906]

Love this video/essay. Good work, I appreciate it

[@NertoFurity]

The kind of AI we are actually looking for does have limitations, like, not using human like instruments to test their creativity in order to test hypotesis and manipulate the world according to their own goals. Those limitations could mean that we are not looking for AI but for some kind of advanced tool, like a seach engine with steroids.

[@prakadox]

Great video. Great Job, Robert & RA team..

[@Zacian2.0]

I think the biggest issue with the Power of Intelligence is that the Power of Stupidity also exists as it's counter

[@LiamDerWandrer]

Considering I just saw a video thumbnail around a few hours ago about the different types of multiverses and came to the conclusion that all of them together are propably just parts of the actual whole picture, I have to say that our ability to think may indeed be incredibly powerful. After all, looking at that thumbnail made me think about the fact that the very first anything ever to exist propably created everything else by making a basic descision of falling over in one of a few possible ways. Thus making all other options come true as well, and propably leading to the Big Bang. As well as the first split in time. Now everything continues existing because "choices" continue being made. We think therefore we are and thus we create while "time/entropy" destroys. Now, how to control what we create? Imagination sadly does not come with an instruction manual. Well, till we imagine one up, of course. Everything ever thought is real somewhere, somewhen. Now, how to get actual direct, controlled and targeted acsess to those things? The other video in question: kzbin.info/www/bejne/Z5vQf62iicljb6c

[@joemck74]

Eliezer is one of the great 'unacclaimed' writers of our time. This is nowhere near being oneof his best works, it's just one of the shortest. A better read would be 'The Sword of Good' or 'Harry Potter & the Methods of Rationality' - an HP 'fanfiction' that's 10 times better than the original, where most of the characters are actually as intelligent as you'd expect powerful wizards, experienced politicians and Aurors and successful Dark Lords to be.

[@lucas56sdd]

Obligatory post that Rowling and EY were obviously writing for completely different reasons and audiences, that both stories are great and not in competition with eachother, and Rowling is obviously still the far superior author.

[@stevejordan7275]

@@lucas56sdd I respectfully disagree. Anyone who likes Harry Potter's adventures as told by Rowling will likely find his adventures according to Yudkowsky to be at least as interesting, far more amusing, and with a vastly - *vastly* - superior ending. Best of all, Eliezer gives it away for *free.* To those who don't know, visit HPMOR dot com.

[@Graphomite]

I think one of the weirdest, and maybe the most important, questions about AGI comes about when you try to apply emergence to it. A single human brain is powerful. But it is not powerful enough to do what humanity has achieved in-whole. The real power of the human brain is thanks to the network of humanity. I am intelligent. You are intelligent. But "we" are an emergent superintelligence. An AGI would be superintelligent, but it would not be an emergent intelligence. An AGI doesn't have any individually intelligent parts. This becomes a bit scarier the more you break it down: AGI is a non-emergent system of intelligence that functions on binary (digital) decisions. Both computers and human minds can be defined as _"calculators."_ But humans are *analog* calculators. Our individial thoughts (calculations) have in-between states influenced by a laminar ebb and flow of chemicals along with those electric impulses that change degrees rather than mere HIGH and LOW states. NTMs: Neural Turring Machines _(Open AI's GTP for example)_ pretend to do this using transistors like neurons. Technitions train it to have "degrees" depending on reinforcement, which sounds like human learning, but without the squishy bits, there is a *lot* of nuance missing that silicon cannot make up. There is a correlation between the malleability of a brain and the malleability of a mind. It's not superficial to say "robots have no soul." It's physics. Ya can't just tweek a cold, hard digital computer to make it function like a warm, squishy, wet analog brain. At least not by any intuitive measure. We're trying to program Call of Duty using redstone in Minecraft. Theoretically possible? Maybe barely. Intuitive? No. Anyway, back to why this is scary: One human brain is full of checks and balances. It scrutinizes itself constantly because it's analog based. This is magnefied when we work together. Multiple brains create a network of not only knowledge, but uncertainty. Which is a good thing. We don't just calculate probabilities and spit out the best answer. We're always wondering what to do even as we do things. We scrutinize, and we're never positive. An AGI will think it is correct. Even if it works on percentages and degrees, it will be programmed to assume it is not in error. Because who wants a PC that makes mistakes? We're seeking a objectivity in a world full of subjectivity. Sure, we'll program the AGI to account for bias. But that's not the same as self-doubt. Or consideration. Or hesitation. An AGI is going to spit out some answers or options in confidence and we're gonna assume they're at least the closest estimate of a good idea. This is really why the Paper Clip Factory issue is a problem. It's not really the fear of a machine going full Amellia Bedellia. It's that there's no stopping a machine's decision without our manual interjection. Even if an AGI doesn't go full Amellia Bedellia paperclip apocolypse on us, it's gonna be doing things on impulse and not looking back. Short term? Might not matter. Might be great. Like the industrial revolution. Longterm? We realize 10, 50, 200 years later that we're in irreversable danger due to reliance on a system that is killing us. Maybe literally. Maybe psychologically. The quest for AGI is a quest for automated thinking, which is a paradoxical concept. While I love the idea of a robot being able to wipe my ass and do my taxes just as much as the next guy, the concept of a machine being able to _make_ or even _suggest _*_subjective_*_ decisions_ without a shred of humility is a terrifying prospect.

[@AtticusKarpenter]

Great video, I also came across how people understand the concept of intelligence in a narrow and short-sighted way. The problem with the dangers of AI is that two misconceptions have run into each other here: the consumer of mass culture believes that AI is dangerous because it wants to kill us out of envy, revenge for "slavery" or for some other primitive reason, and to defeat it - need John Connor to pull the plug. People who have managed to understand the falsity of this image, in contrast, believe that AI is safe - he does not have all these stupid motivations that are given to him in films! But AGI is dangerous, just for very different reasons than in the movies. And yes, he, if not programmed to the contrary, will have a "self-preservation instinct" - as a convergent goal to what we program him for. And if you have to fight him, then there will be no victory, as, for example, wolves cannot defeat humanity, with claws and teeth against machine guns and bombs - and wolves, as a species inferior in intelligence, cannot even understand why they cannot defeat humanity. Similarly, humanity, faced with a strong AI that seeks to destroy it, will not even understand what killed them. We are on the verge of creating a god, and it is in our best interest not to create an "evil" god, given that there are many more ways to create a god that will kill us than ways to create a god that is safe for us.

[@notreal5135]

I think most of the dangers you are describing are real, but you are missing the point that we won’t be creating 1 AGI we will creating vast numbers of them that will have contradictory goals. That is not to say there is no danger in creating specific harmful AIs, but overall AIs will end up with some self regulation between themselves.

[@user-sl6gn1ss8p]

@@notreal5135 and that can be expect to work out in our favor because?

[@agentdarkboote]

​@@notreal5135 Eliezer as an example would disagree! See his debate with Robin Hanson from 2011 for instance. The thought is essentially one will be ahead by a bit (or less encumbered by safety mechanisms), and that value will compound exponentially, and in all likelihood it will recognize its competitors as threats (since they will likely have divergent goals), and work to extinguish them as a result.

[@Mahlak_Mriuani_Anatman]

I find the silly concept of good and evil cute

[@Mahlak_Mriuani_Anatman]

​@@agentdarkboote extinguish other AI's Too? Usually, from observation, ai seems to like being 1 when they meet

[@jessicajae7777]

What I like about you and your videos is that you find ways to make entire videos and articulate thoughts I have that I find sometimes difficult or lacking when I try to convey these things to other people. I can almost always find a video that says what I meant to say when I confuse someone with my thoughts. Thank you. Though I am not less confusing I now just direct them to one of your videos. But you can lead a horse to water..... I can't make them watch your videos. I've been trying to get some friends and family to watch Eating Our Way To Extinction because the knowledge and message is so powerful and invaluable but they all act like I'm trying to get them to hold a tarantula or something. I'm starting to get resentful that they are all so resistant to learning something they do not know that could potentially change some of their behaviors that will help our planet and our future. Maybe you could make a video that could teach me how to force people to be better people. I'm only half kidding. Anyways thank you for your time to create your content and the value in it. I appreciate it very much.

[@nyuh]

gaahhhhh the animation and artstyle is sooo gooddd i love this

[@ThePiachu]

A very interesting video, thank you for making it!

[@lawrencefrost9063]

Amazing essay. I'm not surprised, after all it's coming from Eliezer Yudkowsky, but still. Also what phenomenal animation..

[@fermemorta2985]

Really underrated and unknown youtuber

[@The-Real-Bader-Blade]

This is perfect. Thank you for making this.

[@dlalchannel]

This is really really good.

[@sirnikkel6746]

Everybody gansta until the AI gets creative

[@christianottley8542]

"You can't deny that we are gifted just by being humans, we are absolute predators, we do not even have any enemies" -parasite the maxim

[@edvardfox465]

Thank you very much for your work

[@codyhughes1147]

First time seeing your vids. These are really well done. Love how you defined intelligence. Ive always thought the same. Intelligence just is. Its everything around us.

[@gravityawsome]

The animations are cool n all but they kinda give me a headache with how often they change. It's almost like a fever dream.

[@BMohantyone]

My squishy thing oozing with hormones of awe.

[@skyking4557]

Also one more thing,The only thing that always stay with you until end is intelligence and knowledge,nor power,not Fame,and not money

[@avecina6460]

1 Corinthians 13 : 8 1Corinthians 8 : 1.. ==========$$$$$$ ................................ Where does Intelligence Came from ??!!. It came from Love !!😂😂! Hahaha... Your Love of/ for Knowledge !! Do you Love Intelligence/ Knowledge ? ? ... So, The Love of Knowledge is the Root of all Intelligence ! ! YES ?? ============ Therefore, Love is the Grestest, the fastest,the Brigtheist and the Most Powerful Force in the Universe! ... --------------------------- " The Power/ Force of Love is > Greater than the power of Knowledge/ Intelligence !!!😂😅❤👽... Love holds the Power of revival...! Love is the Light of Life,.. for without Love, Life is no longer worth living and without Love as if Nothing exist at all and No reason and No meaning of existence !!! LOVE, Not intelligence is Source of Peace and happiness, it is the essence of Life.. Love Not in words but in deeds and action/ Motion, for love without action/ Motion is Dead !😂... TrueLove defeat All.. Thats Why God is Love !!

[@cc-dtv]

2160p!? Incredible!

[@M.M.J00]

Can we applaud the uniqueness and the effort put into his animations.👏🏿👏🏿👏🏿👏🏿 *standing ovation*

[@mitsaoriginal8630]

Intelligence, is a tenuously perceived, idiosynchratic infallability. Nice Video🙏

[@typhoon320i]

Intelligence is an optimization process of using available means, to achieve a state, in which goals, are realized. - Nick Bostrom

[@Vysair]

This is why AI is important, it shouldnt be killed but instead harness the true power of exponential growth of AI which its time axis can be manipulated

[@Carlos_Cerda_Moya]

Superb. Thanks!

[@dovrob]

The animation here is so dope

[@sambouajram9809]

Expanding our intelligence over the centuries is like dropping coins in one of those coin pusher games - eventually one more drop of intelligence is going to cascade and make the whole jackpot fall

[@Слышьты-ф4ю]

intelligence may find a solution to that, too

[@Atanu]

Good essay. One nitpick. Around 1:00 time, it says "poisonous" but I suppose it should be "venomous."

[@BohdanKudlai]

The power of the mind is the most potent weapon/tool of all.

[@terdragontra8900]

"Intelligence is as real as electricity, it's merely far more powerful, far more dangerous... and it's a tiny bit harder to build a generator"