Nick Bostrom on the Joe Rogan Podcast Conversation About the Simulation | AI Podcast Clips
ฝัง
- เผยแพร่เมื่อ 26 มี.ค. 2020
- Full episode with Nick Bostrom (Mar 2020): • Nick Bostrom: Simulati...
Clips channel (Lex Clips): / lexclips
Main channel (Lex Fridman): / lexfridman
(more links below)
Podcast full episodes playlist:
• Lex Fridman Podcast
Podcasts clips playlist:
• Lex Fridman Podcast Clips
Podcast website:
lexfridman.com/ai
Podcast on Apple Podcasts (iTunes):
apple.co/2lwqZIr
Podcast on Spotify:
spoti.fi/2nEwCF8
Podcast RSS:
lexfridman.com/category/ai/feed/
Nick Bostrom is a philosopher at University of Oxford and the director of the Future of Humanity Institute. He has worked on fascinating and important ideas in existential risks, simulation hypothesis, human enhancement ethics, and the risks of superintelligent AI systems, including in his book Superintelligence. I can see talking to Nick multiple times on this podcast, many hours each time, but we have to start somewhere.
Subscribe to this TH-cam channel or connect on:
- Twitter: / lexfridman
- LinkedIn: / lexfridman
- Facebook: / lexfridman
- Instagram: / lexfridman
- Medium: / lexfridman
- Support on Patreon: / lexfridman - วิทยาศาสตร์และเทคโนโลยี
If I am living in a simulation - I would like to be transferred to a better one.
That is the very point - your suffering is meaningless to the creators because it is simulated. As opposed to an all powerful god who tortures you for its own entertainment. I find it kind of comforting.
Transfer accepted. We are experiencing a lot of traffic for transfer requests at this time, the estimated time for transfer is 50 years. Your transfer is very important to us, thank you for trusting Earth with your experience.
Make it better
The rich have figured out the algorithm to achieve higher levels.
@@PersonalityMalfunction
---What? What's the difference in the suffering. I mean there's a main difference in power dynamics as even "gods" may have "gods". There's comfort there.
Probably spend more time on achieving something constructive and maybe someday you'll wake up somewhere else.
watching an interview with someone about his interview on the joe rogan podcast man that podcast has power
@@noomade It was like Joe depreciated Bostrom based on his lack of athletic skills.
@@noomade we all have our moments. But how many times does Joe have to say he is a dumb knuckle dragging gorilla before you are shook like this
A type of simulation...😊
Lex rolled his eyes when Nick said "I call it the "bland principle of indifference" lol
Beautiful explanation and discussion of the theory
This is both fascinating and frustrating. I ❤️ it!!!
mind bending and fun. thanks!
the complexity of the simulation may have to do with the graininess at the planck length, that is something that intent affects. the quantum landscape like the shape of every mitochondria has uniqueness that is a point on the vacuum is unique compared to most all others, to sync such quantum landscapes may affect the complexity of the sim
the microtubles are unique as well, except when in bulk i assume, these may somehow synch nonlocally with the construct, or the background, or that is to say the vacuum. i wonder if there are black holes at the smallest length....
how to fry brain :
put some oil on your head
watch this video 3x
ad salt and peper
serve when hot
It was painful listening to that part of the podcast...
but still love Joe.
Man.....agonizing to watch. Sometimes he starts to get aggressive and keeps repeating himself over and over again despite it being explained in many different ways. It seems like he’s simply unwilling to accept that maybe every variable of his life is not actually real.
@@cuculan1978 you can’t, it’s just a matter of considering the POSSIBILITY that life is not what we think it is. Joe is unwilling to do that.
@@granthoover9045 but assuming that we are living someone's computer is silly. Even if it is true, it still doesnt explain the true nature of reality. And it still doesnt explain why, an ancient civilization would invest this much energy on simulating a universe that Im pretty sure they already understand
@@JBSCORNERL8 “it’s silly to think humans would make video games. They have real life. What would be the purpose?”
@@granthoover9045 we create video games for entertainment and It doesn’t cost us many resources. You’re talking about a whole simulated universe which would take enormous mounts of processing power.
thank you
I have question about multiple sims. Assume number of "incesption" simulations one above another. is there some kind of losing information or quality because of grand sim tiking speed of clock and amount of memory? Every another level down it would be like the film inception where everything happens slower and worse quality of lower ammount of memory, but not for people whose there? They never know how fast there lives going? But maybe hope is in this quality deminishing. I've ask to fast:D. Dr. Lex asking this in 20:00.
NB should direct a Black Mirror episode
In ST, is it necessary that the objects of interest that are simulated are humans?
Is lex speak about DMT
If the simulation procedurally generates reality in space and time dimensions within the frustum and occlusion/LOD levels of the observer then in principle it wouldn't need a cosmological amount of matter/energy to run the simulation. It would be like a sort of library of babel. Even our minds could be part of the procedural algorithm and free will be just an illusion, there would be no free choices to make, no randomness or chance. It would all follow the next step of the procedural generator. Even if we don't live in a simulation this could also be the case if the universe is deterministic.
Random personal theory that ties in a bit:
I've been debating the properties of the separate hemispheres of the brain due to the studies on split brain patients...
In short, whenever the left (language comprehension) takes over the right (mute), pushing aside its input-what if the other entity inside is actually an observer that intended to live a simulated life from a human's perspective, but the AI that makes our programming automatically takes over (having developed what we call "free will") until they "die" in the simulation (our life).
If we are a simulation, imagine the entropy cost.
Think im more confused now lol
I dont know how many people are me presently or in my hive, but i can tell you that the outcome of rolling dice, and flipping to highly correlated words and sentences to thoughts is known aforethought, before even opening a book, before even rolling the dice. i can also tell you that the packets of information sent by the machine can be known before the machine has even formed the packets... they can also be amanipulted before the machine sends the packtes, go figure that one out :-)
This is when math starts to break
It makes me think of determinism. Why/how would a simulation of consciousness be apt?
Copy a brain. Now you have a computer that runs a conscious program
I think there is a 4th scenarios; it is not possible to create concious beings in a simulation. Nick Bostrom builds his whole argument chain on the premise of substrate independence of conciousness, meaning that the mind doesn´t need a brain necessarily. I think we just don´t know enough yet about the brain and conciousness to assume a complete substrate independency. Therefore building a logical chain is kind of redundant, when you can not prove the premise yet.
One thing that seems missed in this simulated world idea is the question of resources required to run a simulation of an entire world. How many atoms of silicon and how many electrons of electricity are required to run a simulation of an entire world? Is there enough matter in the universe to run a single simulation that is as complex as the universe itself? Sure, if space is infinite, you could theoretically pull together the resources to do it, but how likely is it that all of those resources would be present in one spot.
Or, do you just have to assume that our simulation may seem complex, but the outside world could be more complex? Do we assume that the world outside of the simulation has different laws of physics that allow for simulating complex things in tiny spaces? Where is my logic off here?
Wait, no, he addresses it in the last half.
Is this guy on prosaic?
Joe "I refuse to accept that we are living in a simulation" Rogan.
What about energy ?
A parent simulator donate less energy to next generation simulation and so on...
Bro this is intense think about it wtf
The qualia are real by definition. Beyond that I've come to realise there isn't any point in speculating.
MJO "If we could look into a telescope or microscope , to the the big bang to birth, we would see ourselves. " we don't die.
So a deterministic simulation, if we can eventually create a simulation ourselves, shouldn't we be able to recreate our universe and if so, aren't we facing paradoxes and alot of stuff by being able to look into the future?
Jonatan arvidsson Not necessarily. If you interpret quantum mechanics using Everett‘s Many Worlds you might be fine. It‘s the “same“ as with time travel.
@@14u73 Well, you do have a ton of math paradoxes if they run deterministic and happen the way our universe does, HOWEVER, in theory it could be that this machine can show many future universes, from where we are, with our past branches but it would be impossible for us to project the correct future universe with our machine, or something.
Joe "Huh?" Rogan
If ts a stimulation then it would be constructed in a computer and the observers would be looking at compresed data. I can't see a reason to have both present in the simulation as you could only observe in real time if you where in it yet you could speed it up and analyse it later
If it is a “stimulation,” hopefully there is a happy ending.😃
If we get to the point where we could start a simulation wouldn't self preservation be a reason not to start it. Because if we were in a simulation ourselves, turning on an other ones would raise the computing power needed for the simulation we live in and it might get shut down or at least take more time to compute?
2 things: 1) what if the original simulator allowed it because they are aware of their computing capabilities? 2) the first simulation that advanced enough to create a second level simulation might not know they are the first, and perhaps it necessarily will only manifest a second tier simulation once the CPU reached the necessary ba dwidth or computational power.
With today crisis I feel like I have some kind of strange retention regarding all scientific speculation and theories, like maybe I shouldn't be this out of this context...
That's a good thing, as long as you know what IS speculation, and what IS supported.
Is this is a podcast based off of a conversation that a podcaster had with this guy on his own podcast?
it's based on conversation he had on joe rogan podcast with Joe.
The discussion here seems to make the assumption that all the simulations are running with the same laws of physics. There's no reason why a simulation would need to use the same cosmological constants that ours does. Who knows what the lower level simulation actually looks like.
Our reality could be working slower than the next level, we wouldn't know it, therefore resources could be much less constrained than we think.
How does this affect the probability that we're living in a simulation? I can't think of any arguments that were made which your comment is relevant to
I doubt "slower" can possibly have any meaningful definition in this case. Slower compared to what? Design an observable race.
It's covered by the first condition of Boströms Trilemma
This universe is so ancient, that chances are, out there within the void, many other civilizations have already long ago evolved and advanced so far beyond us, that they were able to create ultra complex and elaborate reality simulations and so we are most likely living within one of them now.
Unless you accept the idea of the “great filter,” which increasingly intrigues me.
@@dashx1103 what's that about?
@@michaelbruns449 It's a belief on the part of some scientists and philosophers that life, even intelligent life (anywhere in the universe), will necessarily fail to reach the point of things like interplanetary migration. Part of the thesis for most is that technologically advanced civilizations will necessarily destroy themselves before reaching yet to be observed milestones. Before becoming "posthuman" is frequently used in discussing it regarding our little corner of the universe. As a jump off point, I recommend Nick Bostrom's paper: "WHERE ARE THEY? WHY I HOPE THE SEARCH FOR EXTRATERRESTRIAL LIFE FINDS NOTHING." (I tried to link to it, but that negated my post.)
By the way, just to be clear, I agree with your original comment. I was just adding on to possibilities. And Bostrom has written a lot about that too -- including his famous (infamous?) paper: "ARE WE LIVING IN A SIMULATION?" That idea, the simulation trilemma discussed in this segment, intrigues me as well.
Would probability suggest that we are more likely to be in a very low resolution simulation?
Only if you accept a couple of very big assumptions as being true.
PascalsWager5 - No, the opposite. If we are in a simulation it is likely we are in one of the most highly advanced and high resolution simulations. Take the analogy of technology such as film for example. If you count all the experiences of people watching movies since movies were invented and pick one experience at random, do you think it is more likely to be some old black and white silent movie or some recently made film? There is an overwhelming probability it will be a modern movie because the number of movies made grew exponentially together with the quality of film technology.
Simulated worlds would likely follow a similar pattern. Assuming we’re in a simulation we’re much more likely in some super-advanced ancestor simulation being run on a planet sized quantum computer millions of years in the future rather than some half-assed early brain in a jar prototype.
@@sparky4747 but isn´t the assumption of many simulations going on based on the argument, that you have simulations in a simulation in a simulation etc.. Wouldn´t this lower the resolution of each new simulation, because each one embedded in a previous simulation must always be lower resolution than it´s "host simulation"?
So if you cant prove your in a simulation then your in a simulation.
That makes zero sense
@@JBSCORNERL8 can you?
@@emmanuelboakye1124 thats like saying god exists aslong as no one can prove he doesnt exist. Plus, this assumes thst an alien civilization would even want to simulate the universe. Not to mention the amount of processing power to do it, is insane.
@@JBSCORNERL8 i get you but it is likely if you think about,we have simulations now,imagine if they get powerfuller.50 years from now we will see.
@@emmanuelboakye1124 I think its possible but Im not sold. I think its alot easier to look for a theory where the universe simulates itself based on mathematics being fundamental to our existence. Humans always are looking for a creator and that only makes the problem more difficult.
Several flaws with this line of thinking assuming that each simulation is creating a simulation with the exact same laws of physics as the original reality. If we follow the probability argument, there would be no reason as to why this would remain a constant throughout each simulation. A different approach one should consider is that the laws of physics set in our existence is vastly more defined and constricted than the first version. Going further in this theory, one can remove the boundaries of what is possible in terms of physics and the types and amount of resources in previous simulations.
I’m missing the JRE part. Clickbait? I already watched this on your channel.
He is talking ABOUT what he talked about in JRE
You would know and beable to figure out, even with life times
Figure out what? That we are in a simulation? Do you think your little character on Call of Duty thinks he’s in a simulation? Lol In Elon Musk’s words, if you just see where we went from something as simple as the game pong to where we are now and have virtual reality games, in a hundred years or even a thousand years, even if we don’t have a significant jump in technology and just continue to go at the pace we are now it is almost impossible that we will not have the technology to be able to create a simulation comparable to what we think real life is right now!
@@stephenyoho9338 i dont believe in human existing simulators unless your a robot of some kind lol
@@stephenyoho9338 No it’s not impossible to think that we might not be able to create consciousness simulated in some kind of digital environment. The simulation hypothesis is not a video game or some virtual reality that people in the “real universe” are playing. It’s saying that somehow civilizations created conscious beings that exist in a simulation.
We don’t even understand what consciousness is. Also there’s no way to EVER tell if artificial intelligence or a simulated brain is actually conscious. We could(and probably will) create some incredibly advanced android who could respond in all the appropriate ways to mimic a conscious being, but there is NEVER a way to tell if this being actually has the subjective awareness of experience that we call consciousness.
And how do you avoid the problem of an infinite regress of simulations. If you follow his logic then the simulations will create simulations who in turn create simulations and you’ll have a never ending infinite regression of simulations.
And what is the motivation for the simulations. Are we being watched like some kind of entertainment for them. It’s such an anti-scientific hypothesis because ultimately what it saying is there’s some kind of external reality that may not have the same fundamental laws and we will never be able to access it.
People think they can pick up on certain details that might lead us to think this is a simulation like a glitch or something. But what they’re missing is that if this truly is a simulation then we have no idea what the non simulated universe is like to be able to make a inference like that.
It seems to me the "simulation" hypothesis is similar to the God hypothesis in that it postulates a higher being who created us. Most scientists would agree that there is no evidence to support a God hypothesis, and for that reason it is not further investigated. Would not the same argument effectively remove the simulation hypothesis from further study?
It seems to me that the simulation argument is your answer Ralph. It provides 3 possible scenarios.
Which one do you find the most plausible and why? Or do you see a 4th...?
You should read the whole argument, one of the possibilities is that humans never pass the great filter
The difference is the following: We don't know if we can ever demonstrate that we're in a "simulation". If we can't, then we don't pretend it's true just because we enjoy the idea. I don't think the "simulation" is being "studied" right now. What exactly would they be studying? We have to wait until the answer to this question seems within our grasp, then we study whatever might have evidence for it. How does the God hypothesis follow the same path?
@@zuzusuperfly8363 It follows the same path in the sense that both of them have no scientific evidence for their validity, nor are there tests that can be done to falsify them. E.g it is not science, it is just philosophical speculation. And even if you can gerrymander and contort language to make such an idea sound convincing to the human brain, that doesn't make it true. And the same goes for religion.
PascalsWager5 there definitely is a 4th scenarios; it is not possible to create concious beings in a simulation. Nick Bostrom builds his whole argument chain on the premise of substrate independence of conciousness, meaning that the mind doesn´t need a brain necessarily. I think we jsut don´t know enough yet about the brain and conciousness to assume a complete substrate independency.
I can easily see 2 things wrong with the doomsday argument. First, the same rule has to apply for the person #10 000, for example. If they thought of themselves as a random sample, it would be very unlikely that they would live in a world, where people would reach numbers like 100 billion. Yet, the person #10 000 happened to be indeed in a world, where people actually reached #100 billion. Second, we don't need to reproduce and multiply in this same old biological way forever. We may indeed be some of the last humans to ever be born, because we might discover how to reverse aging and even upload our consciousness into machines, so we can live forever and no need to reproduce.
It seems to me that "The Doomsday Hypothesis" (and similar arguments) only become valid when the set is complete. For instance, if we assume civilisation started in earnest circa 10,000 BCE, then what would someone conclude if they were born in the year 8,000 BCE? They would conclude that civilisation only has 2,000 years of life left and would end circa 6000 BCE. So much for "The Doomsday Hypothesis".
I'm dumb.
Why does Lex sound like he's smoked 10 cones?
The problem I see with the doomsday argument is that in the example with the urns, both urns have reached their final value, while the human population is still growing, our urns are still being filled.
So if we say that to fill those urns we need (hundreds of) thousands of years to reach the final number of balls and assume both urns are being filled at the same rate, if we then at year 5000 take a random sampling from one of the urns, they will actually both contain the same number of balls and we'd still have a 50/50 probability.
Still learning to resist the woman in a red dress
#BasementReality
Paul, man i LOVED that ending. now i gotta go look this dweeb up. thanks for all the knowledge you shared i have learned alot and been reminded of more that i had forgotten.
Chronic meth users would love this subject.
This is just modern day intelligent design
You are all wrong the simulation isn't on us but rather on how bacteria and viruses evolve hence why were having a pandemic rn.
This is such a better interview then Joe did. As a matter of fact that was Joe's worst job as a interviewer that I remember. I was so surprised Joe Did not get it. Dmt and the simulation is all Joe talks about if it's not animal kills, hunting and eating meat.
I bet u still don't get it.
@@the3es48 well if its anything good then I am sure I don't get it. 😂
@@the3es48 O sorry if I touched a nerve for those who did not get it. 😭
Simulation theory is nonsense. Proponents are caught up in their own metaphor
No less nonsensical than any religion
There is a difference between not personally believing something and it not making sense.
His argument does make sense.
1st
This guy has hair like me.
Could we not, in principal, using quantum physics and crucially without collapsing wave functions, create an infinite set of simulations albeit with a particular set of parameters?
OH SHIT, did i just become God? fucking sweet :D haha
Although i suppose it may be the case that none "happen" without collapse and collapse would result in one simulation...hmm..unsure
Complete wilderness, simulation my arse 😶 a simulated smack to the face would not hurt, whereas a real one will. Please feel free to volunteer for a real one and I will happily oblige.
Shaun mr coolguy
Great! First
Never trust anyone who has overactive jazz hands.
200th
If your one of the lucky 60k that watched this. You’re blessed
He fails to address the possibility that it might be impossible to create a simulation as complex as our universe no matter how advanced a civilization is.
Why?
Seems a little click batish to show Joe and Nick and put a matrix background but it's just a clip from a Lex podcast... That's like putting Darth Vader and Han Solo on the cover of Blade runner 🤔
Is the Simulation just intellectual diarrhoea ?
You can work out the probability that it is.. but it will take a lot of debate and calculation :-)
Two things strike me about all of these kinds of thought experiments and intellectual inventing problems to solve that don't exist. 1. Why would anyone want to simulate our entire Universe? 2. Why would the fact that it is a simulation be hidden from everyone?
Something about Nick's explanations and justifications is a red flag also that he seems incapable of a simple straight-forward explanation that seems coherent. The entire basis of his reasoning is that probability suggests anything is possible. I think that's his strongest position and everything else he says weakens that.
@John Samson yup
Why even care? For all we know Iguanas are sensory extremities of a vast 5th dimensional being with similar sensory extremities dispersed through this and other universes.
For all we know we are just version 684.5B of said sensory extremities. And yet... Even so we would still human. Still on earth. Still doing our thing the way we always have.
Reality is subjective, and beyond our subjectively perceivable reality is more than a universe worth of possibilities.
Just do what you enjoy doing.
We don't even have enough of a handle on the natural laws to make a good guess anyway. Maybe once we can monitor individual quarks. Who the hell knows? Whatever else. I'm not putting God in the blank space. For lack of evidence if no other reason.
I agree. The problem with people is they get stuck worrying about money, competition, being better than the next guy etc. It's a sad existence. I will take your advice and enjoy life.
@Raccy They should be having this debate separately in soundproof rooms. It would do about as much good.
this is not a wise assumption
Sounds like a lot of BS to me. Seek God,
If show me your intelligence level was a comment
This is both fascinating and frustrating. I ❤ it!!!