The Path to SuperIntelligence & Escaping the Simulation — With Nick Bostrum
Channel: Alex Kantrowitz
Published at: 2024-08-07
YouTube video id: a4SyxHnDhgE
Source: https://www.youtube.com/watch?v=a4SyxHnDhgE
why do you believe that it's a certainty that will sort of achieve artificial general intelligence or super inell really certain certainty is too strong a word I think you feel it's very likely yeah at least conditional on um Science and Technology continuing and like we don't don't we could have some sort of civilizational collapse or we could eventually um you know go extinct or destroy ourselves in some other way not unrelated to AI there are like developments in synthetic biology and other new weapon systems and whatnot um but if if kind of these development efforts in in in in hardware and in in in algorithms continue then it looks very likely that we will succeed in this I mean from first principles we have the human brain as an existence proof that general intelligence is possible there is no reason at all to suppose the human brain is in any sense optimal like neither from a hardware point of view uh nor presumably from the algorithmic point of view um so just as we have machines that are physically much stronger and faster than human bodies or any animal body is then likewise we will have eventually cognitive systems that are much faster and clever um so that that's like a very high level um argument but then you can also look just at the kind of advances we are seeing where more and more things that used to be impossible for to do have become done uh and um there are just a lot fewer left of these kind of uh Milestones um there is this phenomenon that like before it is done it looks really hard once a I have done it then we kind of quickly forget how just how like impressive it was and just take it for granted like yeah of course computers can play chess like of course what's the big deal but like at the time like it was a big and then you can see like oh we have they can play go they can see they can imagine they can write poetry they can write computer programs they can talk to us in ordinary language they like can pass like you know um undergraduate level exams in all these different subjects like this this this is a lot of stuff that like to people 30 years ago would have been like wow uh you must be really close to ADI if you have done all of these things you know there's a philosophy in Silicon Valley and some people dismiss it but I think you know they should give it Credence because a lot of the money is behind it that it's called effective accelerationism it's almost like a meme on on Twitter and um you know I'm curious what you think basically it says we should just continue to develop and not pay any mind to the potential risks because ultimately just the active building is good in in and of itself no I I think we should pay heed to the uh risks um um um yeah people have a tendency to sort of simplify a complex set of considerations down to something you could write on a banner and then you form a little tribe that that kind of carries that Banner who who then has an excuse to sort of uh fight verbal battles against people from another tribe to carry another banner and it seems like a very in inefficient way for human Collective deliberation to occur that there are these kind of like semi idiotic tribes that kind of hopefully in aggregate kind of represent all the different views and may maybe it sort of shakes out so that the relevant considerations are taken into account but um it's almost like each person to the extent that they join the fry in that way like reduces themselves to being one neuron in in this Collective brain and then hoping that it adds up to sanity whereas I think our Collective brain is is kind of a bit uh crazy sometimes right I'll continue no um so yeah I I think some of it is also more like a kind of FR frustration with the ways that um other forms of technological progress uh is is styed by various culture like kind of overregulation for example like where each little regulation that some some government or the EU passes seem like in isolation kind of well you know it fixes this problem but then you end up in the situation where you have like millions of regulations and the overall effect is this massive drag on on just doing things that could help and eventually you get stasis and so and and also like in um amongst people talking about like in like the ethical uh concerns you see it in biom medicine and you see it in other fields environmentalism so there's often like a legitimate problem there that people are concerned with but sometimes people who professionally take it as their job to identify ethical problems like become very good at pointing out what all the reasons are for not doing something all the problems and issues but they kind of get it's harder to sometimes see the the cost of of limiting ourselves in all those ways because the new innovations that would be unlocked are not yet there where it's like the pollution is is there immediately right or the the person like if a self-driving car runs over a pedestrian like that's immediately visible but all the people who could be saved if self-driving is developed and works you know there is no new story about somebody who didn't die from a traffic accident and so if you generalize that across all the different uh sectors of the economy and Science and Technology then it's natural to build up this sense of frustration that oh there are these these people who are sort of strangling the human spirit and like all these good things that we could have had if it weren't for these naysayers and and then you build up like a kind of uh movement to push against that I think some some of the impetus for Effective accelerationism comes from that kind of cumulative frustration and then it's applied to AI in particular and then I mean what they're pushing back against is effectively I mean effective altruism right that's the one that they view themselves as the opponent of and I mean it's effective alism emerged from Oxford and you've I think you're ideas have inspired a lot of their views their thinking on artificial intelligence so how have you viewed the evolution of EA yeah so I mean i' I've never U described myself as an effective altruism I think my uh uh uh some of my ideas have had an in an influence there and in particular um yeah try to think more about the sort of the the macro strategic aspects of our strivings as as opposed to just the local mediate effects of what we do um and so yeah I think there is like a contingent things because AI has emerged as a sort of a very Central cause area for many in the effective alism movement because they now recognize that AI will be a big very impactful thing so trying to shape that for the better seems like really important in this day and age right um and then more recently a lot of that has done taking the form of sort of pessimism about the likelihood of the alignment problem being solved and and hence then resorting to the idea well if we can't solve the alignment maybe at least we could slow down the AI progress so that you know we have more time um and then slowing down AI progress becomes then more of sort of political like it's not like a technical solution you just like figure some somebody's really ever and write something on the Whiteboard and you have the formula right this is more like something that then requires activism um and so and but once you start to engage in activism then there is kind of mtic pressures to to simplify our message and to close ranks and to try to beat the people with opposing views down in the marketplace of ideas um and I think we're seeing some of the beginnings of that um where there are these kind of campaigns uh MH to try to push a certain view um not not like a kind of open-minded intellectual inquiry to try to figure out where the truth lies like there's like an assumption we've already figured out the truth and now the like the the the task is more to sort of uh ramage down the throat of of people who have a different View and and similarly on the opposite side so but this is not Universal there's like I think you effective altruism is is like a broad tent and um um there's like a range of different views in in that field and I I wouldn't want to sort of um paint the whole field with a uniform color there U are a lot of uh people who are very much doing their own thinking and have very complex uh like World models of how how this might play out right okay let's talk about Utopia a little bit um just give us your perspective on what could go right in the besta well why don't we do this way what could go wrong in the worst case scenario of AI what could go right in the best case scenario of AI and how do how do humans have a uh influence in terms of which direction we go well I think I mean at the minimum on the downside I think existential risks certainly are part of that picture and as as well on and there's like a different community so you mentioned like the effective accelerationists and the doomers so those are certainly there then there are also the people who are more concerned about sort of the immediate impact of AI on society um yes you know discrimination or or censorship and IP like and those are also legitimate issue I just want to acknowledge those even though they are kind of like a separate node in this but yeah um so I think there is like the uh the real X risks existential risks that will arise as we develop uh and possibly not even super intelligence but you could imagine even something short of that making it very easy to develop new weapons of mass destruction in using synthetic biology or other so so are you more concerned about humans using AI to hurt each other or AI hurting us um well I think they are both uh worth worrying about um I I think with the X risks I mean maybe a slightly larger on the AI uh being the kind of agentic part there um um but but certainly uh we we really need to do a decent job on both uh sort of the alignment and and the governance for for us to have a good outcome so so yeah so now on the upside um like there's a huge unlock and and a lot of that is just removal of a bunch of negatives um um if if you look around at at the the world as it is now it's not that Rosy a picture in many ways it's quite a horror show uh with people dying from you know Alzheimer's or kids getting cancer and like starvation and people being bombed and like all kinds of or just at the more mundane level people spending most of their uh adult life you know working uh in in a sort of boring occupation that gives them no fulfillment but they just have to do it to you know to pay the rent and like you know headaches and stomach aches and like all all kinds of just the totality of all of this extreme misery and very common more everyday misery um that's just within the human sphere and then you add the animal kingdom with like um so just kind of ameliorating all that suffering would already be I think a very strong argument that something at some point needs to be done here can't like go on like this I mean do we want like another you know 10,000 year 100 thousand years of just this um but I think on top of that there is like the potential to also um unlock like new levels of flourishing Beyond those which are possible even under ideal conditions in the current world um that's a lot harder to paint a very concrete picture of because we are sort of limited in our ability to imagine and appreciate just as you know if you imagine like the great ape ancestors of homo sapiens kind of thinking about what could be so good about being you know human uh and so they might like realize a few things like oh we could have banana plantations and have like a lot of bananas and stuff and and that is true we can have a lot of bananas now but there's more to Being Human Than Just unlimited bananas right like we have sort of you know music and poetry and film and humor and romantic love and like all all kinds of stuff science uh so similarly there is probably like if we unlock as it were the a greater space of possible modes of being there are some in there I'm sure that are extremely valuable um in that that I think AI would be the most plausible path toward words realizing MH um and so that's kind of um but if if I then really Dives in and tries to think more specifically about like what would the best possible continuation of a life starting from like our current human starting point look like then there are some quite interesting philosophical questions that that arise and so this this book deep Utopia it's not really an attempt to sort of well before we looked at the downside now let's make the case of how wonderful the upside could be I think the upside could be extremely wonderful but that's not sort of the thrust of the book it's more like let's just look at this what would happen if we actually did succeed in uh creating a solved World um as I call it like where all the Practical problems are already solved or to the extent that there are problems that are not solved they are in a way better dealt with by Advanced AIS and robots than by us um and there is like some aspect of that condition that at least Prima fascia look quite unappealing to our current uh sensibilities we often Define our sense of self-worth on on the idea of of being a contributor like you're you're a bread winner or you make like a positive difference in the lives of your friends or of society at large you like you you bring value to the world um so much of our existence is kind of um constructed within the constraints of various instrumental Necessities that have been with us since you know the dawn of the human species there always been a lot of things that we need to do just to survive um and if you remove all of those um there is at least initially this sense of kind of disorientation or an undermining of of the like we feel like kind of what's the purpose like we would just be these blobs but this is different from what we spoke about the last time you and I were on on the phone this was in 2019 I think you said that um we'd have to find some new source this is from from our conversation that I put in my book you we'd have to find some new sources of self work worth but the in in Disneyland the job of children there is to enjoy the whole thing and Disneyland would be a rather sad place if it weren't for the kids so we you say we would all be like kids in this giant Disney Disneyland maybe one that would be maintained and improved by our AI Machine Tools so effectively that even if we didn't have to do any sort of sustainment work that gets turned over to AI we could be actually quite fulfilled in life yeah so how do you get from there to where you are today well I think that's basically correct but we can distinguish two different uh senses of uh fulfilled um or of having purpose so there is first of all what you might say the subjective sense like the feeling of fulfillment or the feeling of having purpose like this like this the the emotion of being motivated and you're really excited about what you're doing right like that kind of psychological state that certainly you could have that in a in a Sol World in Utopia the utopians could have like extreme levels of motivation and immersion and subjective purpose that's easy that's like a check mark um and more broadly you can go through different possible human values and for some of you can just write off the bat say well yeah sure of course that would be trivially easy to do in Utopia so in this case through the psychological uh engineering techniques that they would have I mean already you could have like imagine a drug without side effects and without addiction potential that that just induced the state of Fascination and motivation we already have simple versions of that like but you could imagine far more sophisticated ways that would give the utopians like very fine grained d control over their mental states and their psychology that would follow from technological maturity right um so that so that that's that's easy now some people however think that there is also a more objective concept of purpose uh where it's not just that you feel motivated but that what you are doing is actually objectively worth doing um um and that that's a little bit uh less obvious to what extent the utopians would have that in as much as at least at at first sight it looks like anything they could do they wouldn't have to do because they could just press a button and the Machine could do it instead except for those those few things that you bring bring up that actually we want humans to do like for instance like ordain a marriage right like that is something or read a poem people might want this potentially yeah they're like yeah so you could automate everything but except there might be certain jobs where the as it were consumer has a direct preference that the job be done by human in which case by as it were almost by definition it's not automatable exactly and this whole idea of a solved world is basically where AI can effectively take care of everything all of our needs all the production and we are in this Utopia because the machines have done all the hard stuff that we don't want to do anymore yeah uh so a solved world like is a condition of technological maturity coupled with whatever uh degree of sort of um so you could imagine like a very dystopian scenario with advanced technology if you have like a sort of totalitarian despotic right but imagine also that to whatever extent governance problems can be solved uh they have been solved I know maybe you can't solve governance but to whatever extent they are like better and worse in terms of social political structures imagine like we get something at the good end of that combined with technological maturity then that's basically uh the definition of ass solid World um sorry I know I took you on a bit of a tangent there no so so yeah so like they kind of layer so you could say first well you could have like like a simple Utopia might be a kind of post scarcity Utopia where we just have abundance of material Goods so we already if you're fortunate enough to to live in in like a developed country you know with a decent education Etc you're all already pretty close to that you you might might not be able to have like the ideal yacht of your dreams like there are some limitations but if you sort of plot a line that has like the starting point Hunter gather and the end point is like uh complete post scarcity I think we are more than halfway there um like it's a bigger difference to go from not having enough to eat to having enough to eat than you know to to get like a slightly more advanced version of an iPhone or like mhm like a third house if you already have two houses like diminishing returns so so that's like you could first consider this concept post scarcity Utopia okay so then what's a level of as it were more radical Utopia than that well you could have a post-work Utopia where not just do we have plenty but we don't have to work to produce that plenty so it's not just that like we work all day long and then we have a lot of money and we buy stuff but imagine you had all this plenty and you didn't have to work is slightly more radical conception but not that radical I mean there are already people who you know are born with a trust fund or something and they don't never have to work and they have plenty you know again there's limits to how big their palaces could be but like at least some approximation but I think we can then go further and consider even more radical conceptions so I've already alluded to um there is the post instrumental uh Utopia that you could have where it's not just that we don't have to work like to make money but we also don't have to do any of the other things that we currently have to do for instrumental reasons so if you know if you're Bill Gates you still have to brush your teeth uh you still have to you know do a whole host of things just in your everyday life to get the outcomes you want there's like a limit to how much you can ask uh your assistant to do or that you could you know but but in in this scenario like a lot of those other instrumental reasons we have for doing things would also drop out of the picture with like super Advanced automation technology and then I think there's like a step further than that which is a call it a plastic um Utopia where we also have complete control over ourselves over our own bodies minds and mental States using like Advanced techn biotechnologies or newer Technologies think that's going to we're going to achieve that that's wild yeah but I think if you consider what would be possible at technological maturity which we can at least play some lower bounce on through kind of um um theoretical analysis so we we can sort of estimate what kinds of computational systems could be built we can see what kind of molecular uh Manufacturing Systems are are possible to build in our universe even though we can't currently put them together with the tools we have now we can see that there is like a path there other things like um cures for aging and stuff like that like we don't have them yet but there is no you know no loss of physics prevent people you know from from living indefinitely if you had like repair Technologies etc etc perfect virtual realities perfect like ways to manipulate the brain like that they're better than drugs like there like a whole you could there this kind of in fact like a table in the book that outlines some of the affordances you would have at technological maturity and maybe there will be additional things we haven't yet thought of but at least these and so they would I think enable uh us to instantiate this condition of plasticity where where human nature itself becomes malleable um it's crazy so that means there are further um um questions about purpose so right now like if you didn't have to work for a living like maybe some people would say well you know maybe I would start you know going to the gym War to get fit right like that's uh you can't hire a robot to you know go and run on the treadmill on on your behalf but with plasticity there would be a shortcut to that you could pop a pill that would induce exactly the same physiological anity on the way to there with OIC yeah yeah exactly that's like uh one more step in in that direction um and and so the thing is with super intelligence like you get the telescoping of the future so all these sort of Science Fiction like technologies that maybe we would develop if we had like you know 20,000 years for human scientists who work for it we probably will have a cure for aging and perfect virtual reality and space colonies and all the rest of it right but um all of that could happen very quickly if you have super intelligence doing the research and development so so you get this kind of telescoping of the long term um y but yeah so then then there like a further set of things that currently fill the lives of people uh that we wouldn't need to do um including things we do for fun so maybe some person say well you know if I didn't have to work and uh I had like maybe I would play you know some play golf all day long like because why why would you play golf all along well because it's fun it gives me joy let's suppose somebody says that well then in this condition of plasticity that would be a different an easier way for them to get joy they would just pop a pill and they could get exactly the same level of um subjective well-being as as like a a beautifully manicured uh Golf Course could induce and so I can't even imagine that type world that's crazy um yeah so it does then require a fairly fundamental rethink of of what it of means to be human in in this radically transformed condition but it is kind of the Imp implicit uh teoss of our current strivings if you think about so we we try to the little problems come up like we triy to solve it and then like there's like another problem that we all like so oh our food rot let's invent the refrigerator like oh we get fat let's uh like invent o zic like oh our cars pollute let's make cleaner engines like and but if you kind of extrapolate and take all of that to its limit then you would end up in in situation where we can like do everything with no effort right that would be kind of the the limit of Technology um AI you know the goal of AI has all along been not just to automate a few specific things but to provide the technology that allows us to automate all tasks right like AI hasn't really succeeded until all intellectual labor can be done by machine and and um so I think it's kind of we don't we don't think about it like that but if if you sort of see what all of this effort is all these Investments we have in science and technology and our efforts to make the economic system more uh efficient to allow kids to learn more in school like like all of this kind of adds up to some sort of Arrow of attempted progress in a certain direction and you might as well at some point stop to think what happens if we actually get there yes um and then we do end up I think in this condition of a solid world uh and the question is whether we're ready for that so uh Nick are you able to stick around for another couple minutes or do you have a hard I could do a few more minutes maybe okay all right let's take a quick break and come back and ask a little bit about what whether we can handle that perfect world we'll be back right after this and we're back here with Nick Bostrom he's a philosopher the bestselling author of super intelligence and author of the new book deep Utopia life and meaning in a solve world it's out now great book definitely recommend you pick it up um so you you basically said if we get to this perfect world you think right now we're effectively unfit to inhabit it and in your book we sort of look at uh in the early chapters you sort of look at the fact that we've increased our productivity but we using it for consumption rather than Leisure um and that's concerning to you is that part of the reason why you think we're not quite ready for this um so where where where do we fall short in our preparation for Utopia um well I think human nature is kind of forged and evolved under various conditions including conditions of scarcity and condition where there are like instrumental demands on us that we need to exert ourselves make efforts try we need to work just to get by in life this has been true for hundreds of thousands of years it's still true to some extent today although with certain relaxations like for example food is much less of an issue for people living in in in uh wealthy countries and increasingly also for more middle inome countries where obesity is becoming an issue um so there you can already see a little bit of a mismatch like where we're kind of evolved to live under conditions of food scarcity and when that no longer obtains unless we make adjustments like we kind of balloon in size and then we need to try to find fixes for that but I think a much more profound mismatch between where we currently are psychologically and biologically and and our environment could arise if we sort of suddenly moved into a condition of a solved World um so there would need to be some adjustments in that I think scenario if we wanted to take advantage of of all the things that uh would be possible and what would those adjustments be um well I think uh for a start if if we take for example human hedonic uh well-being which might may be like one of the easiest things to look at just like the subjective state of positive AFF effect like are you actually enjoying the present moment does it feel good or or is it like uh unpleasant and so this is like a fairly fundamental dimension of our psychological State some people think it's the only thing that matters if you're a hedonist in the philosophical sense you think that the only bad is suffering and and the only good is is is pleasure not not necessarily in the sense of physical pleasure but in the broad sense of sort of positive mental honic tone um so it looks like we are kind of designed in in a way where we have a fairly powerful habituation mechanism so if somebody's conditions in life improve a lot often you know maybe they win the lottery or something right so they get really happy they win the lottery but very quickly the sort of fonic tone falls back to the Baseline um right because we are not designed for permanent Bliss uh we are designed in such a way that our reward system motivates us to uh produce more effort at at whatever level like no matter how good our our situation is we are designed to always try to want to make it better and so we only get reward when things improve rather than when things are like at a good level to a first approximation it's not completely true I think people under better conditions are slightly happier than people under worse conditions and maybe a lot happier if they're like under really bad conditions it's like certainly has a sort of permanent uh negative effect but now if there were no more opportunities for improvement and and and there is like no need for our Like Instrumental efforts that it seems very stingy that we would still not be able actually thoroughly to enjoy lives so maybe we would want to change that set point so we could all be much happier all the time um to actually relish this this future like it would be sad like if if everything were like super nice and we were still miserable there like and then just living like that for like millions of years and like not really that that that would see to be uh very unfortunate so that that could be one obvious adjustment um um there's like a bunch of other things as well that like you might imagine upgrades of various human capabilities our cognitive abilities our emotional repertoire our ability to connect to other people obviously physical health um and then kind of at the philosophical almost or like um our overall attitude to life like so the idea that you sort of conceive of your selfworth as being based on the uh your ability to make a contribution uh maybe needs to be rethought here like if we can no longer make contributions there there is n risk to that I think there are certain ways in which maybe the utopians could but at least to first like our opportunity to sort of help out other people would be reduced if there is just less misery and need in the world to begin with like if you're a doctor and there was no disease like you need to like find another occupation or you can't base your self worth of being a really good and caring doctor if there's no if nobody's sick like then you have to rethink that and I think at the more General level we would all have to sort of rethink what makes like a human life um have dignity in in this condition where we are no longer really where it's at yeah okay here's the last one it's kind of wacky uh but I thought I'd throw it out there and see what you think if we reach super intelligence um and we hit this Utopia do you think and I know you've spoken a little bit about the potential of us to be living in a simulation so if we reach super intelligence do you there do you think there's a chance that we're going to crack out of the simulation and effectively figure out who's running it if we are in one um that that that that would be one um possible uh scenario right if you're in a simulation the simulation could just end with nothing or it could be rerun or you could like enter a different sort of environment within the simulation continues but the sort of virtual environment changes or indeed you could be sort of uplifted out of the simulation into like the world of the simulators so all of those are at least kind of metaphysically possible conditional on the us being in a simulation in the first place so the the simulation hypothesis expands the this this the space of sort of realistic possibilities and the realist like the space of realistic Futures you might think if you're living just in a simple materialistic Universe you die that's the end your brain rots there's no more experience and there's really not much room for other things in given like the laws of physics and us being purely material with no soul etc etc if you are in a simulation then there's like a much wider range of things that could happen that that would seem perfectly possible given the assumption I mean if there if we're there it would be so crazy to just unzip and poke our heads out behind the thing so yeah okay the book is uh deep Utopia life and meaning in a solv world it's by Nick Bostrom Our Guest today also the author of super intelligence bestselling author Nick Bostrom thanks so much for spending some time with us today I enjoyed our conversation me too all right everybody thanks so much we will see you next time on big technology podcast