SIO410: Futurama and Simulation Theory

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:16 Thomas: Hello and welcome to Serious Inquiries Only. This is episode 410. I'm your host Thomas Smith. Glad you're joining me today. It's been a little bit since the last episode. Couple reasons for that. For one, holiday cheer, Christmas. Celebrate Christmas here and if anyone wishes me happy holidays instead of Merry Christmas, I punch them in the face. No, not really. Just kidding. But if you want to hear more about the silly War on Christmas, make sure to go over to Where There's Woke. On over there on Where There's Woke, but anyway, also it's because I recorded a bunch with astrophysicist Brian Gillis. It ended up turning into like a little bit of a mini series. So it's the kind of thing where I've been like working on the whole thing at once, and then I'm going to get it out to you more at once. So rather than way the recordings went, it made more sense to kind of do it all together and break it up into parts.

And so you're going to get that over the next few days. I'm really excited. Here's the cool series we're doing. Taking a look at the simulation. We're also talking about the question of whether or not we live in a simulation. Now, first person I go to when I want to know answers to deep questions about the universe obviously is Elon Musk. And so, if you want to hear what that guy thinks, Thanks. Here, here's a, maybe I'll let him present a little bit of whether or not we're in a simulation.

0:01:34 Thomas: I

0:01:34 Clip: Given that we're clearly on a trajectory to have games that are indistinguishable from reality, and those games could be played on any set -top box or on a PC or whatever, and there would probably, probably be billions of such computers, set -top boxes. Would seem to follow that the odds that we're in base reality is one in billions. Do you think? Well, I think it's one in millions. Okay. I

0:01:58 Thomas: I found a clip of Neil deGrasse Tyson talking about it that I thought it was funny. In for how the clip ends. It's my favorite.

0:02:05 Clip: Now you're a character in that world and you think you have free will, and say, I want to invent a computer. So you do. Hey, I wanna create a world in my computer. Then that world creates a world and its computer. And then you have simulations all the way down. Now you lay out all these universes and throw a dart. Which of these universes are you most likely to hit? It, the, to argue against the possibility that all of us are not just the creation of some kid in a parent's basement programming up a world for their own entertainment. But and if it's true, what can we do about it? It's like the Truman movie, or the words that we're in. Yeah. Well, he can try to escape by going in the Truman Movie, to go through the barrier. Yeah, but yeah, if you're, if you were programmed by somebody, yeah, no, there's nothing you can. So what difference does it make if I'm programmed by someone? I guess is, I don't know, it I guess it doesn't make any difference at all.

0:02:55 Thomas: This is an idea that, I mean, I was about to say, you know, serious people consider, I guess, Elon Musk. Not a good example of a serious person, but he's certainly an influential person, like it or not. And I don't, I believe that believing that you're in a simulation, I think that's something that has pretty significant effects on how you might think of the world, how you might treat other people, even possibly, maybe, hypothetically.

So I think this is a pretty important question, but set, poisoning the well aside, set that aside, Elon Musk, for a moment. There are very serious people who've put a lot of thought into this. And just as a guy sitting here thinking, it's a very interesting question, we're going to explore that and a lot of related things, very interesting directions that this goes, with astrophysicist Dr. Brian Gillis. The first part, we're going to do a little bit of an homage to Philosophers in Space.

We're going going to do like physicists and a guy in space, because there was a Futurama episode that discussed this issue, presented this issue in a really fun and funny way. And so we're kind of doing a little bit of a Philosophers in space -ish thing on that, talking about this particular Futurrama episode and what it says and how it presents the arguments about whether or not we're in a simulation. So that's what we're going to do here.

And I want to let you know that also there are places, what I'm going to do is probably poke in here and there. It might be just at the beginning and ends of episodes, because some of what we talk about in this series is pretty dense. It's actually a bit hard to understand. So I'm gonna break it off into more digestible pieces. I don't want to throw too much at you.

And if you don't understand something, it's possible I'm going to reiterate or explain it at the end of the episode in a of post script kind of thing, because I do want to make sure we're all following along. Some of this philosophy is, you know, it's complicated. It's hard to wrap your head around. And so I'm going to do my best to make sure we're all coming along for the journey.

So with that note aside, I want to remind everyone that you can avoid the very annoying auto ads that you're about to hear very shortly by going to patreon.com/seriouspod. We also have bonus content, ad -free episodes, and we have a monthly bonus for second -tier patrons and above. Last month, I'm telling you, if you didn't get the house episode, God, that was so funny. I know you heard a preview on the main feed. That was a lot of fun. I want to do more of that.

I kind of want to just do that every time. But this month, we've got something different that will be very fun, so that's at patreon.com/seriouspod. Thank you so much, as we come to the end of a very difficult year for me personally and Lydia and the family. I I cannot thank patrons enough. Truly, I'd be nowhere without you. We'd be absolutely nowhere without you. Thank you so much for supporting the show. With that said, here's an auto ad for everyone else. And then we're going to talk about Futurama and simulation theory. Brian Gillis, welcome back to the show.

0:05:45 Bryan: Yes, hello Thomas. Thank you very much for having me back. How are you doing today? Yeah.

0:05:49 Thomas: Great. I'm excited. We watched a Futurama. I haven't watched that in a while. It used to be one of my absolute favorite shows back in the day. Forgot they re-re-re-did it. I don't know. They brought it back for like what, the third time or something? I feel like it comes and goes.

0:05:55 Bryan: Something.

0:06:06 Bryan: That, yeah, it was canceled, then there was a series of movies, then it was on Comedy Central, and now it's on Hulu. So. I think, yeah. They brought. It back for the third time. Yeah.

0:06:12 Thomas: Oh, okay. Yeah. But it's crazy because you can actually hear the age in the voice. I know it's a cartoon tune so they can stay, they can stay the same age looking forever, but I can actually hear the difference in their age and their voices, because I watched the hell out of that. That show, the original seasons, oh man, loved it so much.

0:06:21 Bryan: Yeah. Yeah. Uh. It's even worse than this, Simpsons, the voice actress for March doing that voice has really. Done a number on her vocal cords. Well.

0:06:30 Thomas: Oh, yeah. Yeah, I pretended that show stopped in like 2000 or so. I think there's been good stuff, but

0:06:43 Bryan: I've. Heard. People say that the last couple of seasons are quote-unquote good again. Not. In the same way as it was before, but they're doing interesting new things with some good episodes, but I haven't checked it out myself.

0:06:44 Thomas: Yeah. Yeah, yeah, yeah, interesting. Well, this one was a fun science-y one, and it was something I think you wanted to talk about anyway, right?

0:06:59 Bryan: Yeah. Today we're talking about an episode from the new season of Futurama. This is season 11, episode 10, called All the Way Down, which was my favorite episode of the new season, which I felt overall the new season was kind of hit or miss. There were really only two episodes from it I really liked. There's this one and the Xmas episode they did. Those are the two I'd really recommend people check out. Are some callbacks and jokes you won't quite catch if you only watch those two, but if a bit iffy on it, I'd recommend at least those two.

0:07:37 Thomas: Was Futurama the one where the Christmas thing was like Santa comes and like kills people or something or like, yeah, that's right. Famous Futurama lore.

0:07:42 Bryan: But Yeah. Yeah, and so Xmas is when everybody hides out. And yeah, Xmas one is a good take on that, which I won't spoil.

0:07:55 Thomas: Okay. Okay. But yeah, this one, you know, it's like, I get it. I don't know why people don't just make new things. Anyway, it's about simulation theory.

0:08:04 Bryan: Yeah. Yeah. So let's talk about this episode. It basically opens up in a simulation and then zooms out. And Professor Farnsworth says he's made a simulation of the entire universe. So there are a lot of jokes in this, just watching the simulation, a bunch of random jokes. But the main plot is kind of dealing with having a simulation of the entire universe and the implications simulations of it so

0:08:31 Thomas: But we're to believe it's a lower quality simulation. So it's just like pixels, you know, like maybe Nintendo original. But

0:08:38 Bryan: Yeah. Like Fry is made up of three pixels.

0:08:41 Thomas: They still seem to have like and inner thoughts and whatever. So, I don't know how much you're saving, you the representation visually be less. But they have to still have all the internal workings to be, like, sentient and stuff. So, that's kind of weird. But, hmm,

0:08:55 Bryan: Yeah. And That's actually A good point, which we can get to in talking about just that's one of the things I wanted to talk about is just how hard it would be to do an entire simulation of the universe but, uh, see, um, when we, uh, get to that, um, now I think I just want to go over the episode to Kind of set the stage for what we're talking about. So first they have this low resolution simulation, but they say the limit on how well they can do it is the power it takes to run the simulation, because the current simulation is very power intensive. As Professor Farnsworth says, it takes more power than 100 bug zappers. So Thomas, want to get your Would would a simulation of the universe that's low resolution take, in reality, take more or less power than 100 bug zappers?

0:09:12 Thomas: Sure.

0:09:42 Thomas: I would think more would be the answer to that.

0:09:45 Bryan: You're right. Yeah, it actually takes a bit more by a factor of 100 or so orders of magnitude. So yes, obviously it takes more. They do this wacky solution using sewage water to power it. They power it up, and they can go to better resolution, which to us, now the simulation looks just as good as reality, except they draw in this payload around all of the characters, just so we know when we're watching the simulation.

0:10:11 Thomas: Just so we can tell the difference, yeah.

0:10:13 Bryan: They do some more simulation jokes, and then all of a sudden the simulation characters come up, and Professor Farnsworth in the simulation says, I've created a simulation of the entire universe, and it zooms down to the same low -resolution simulation we saw before. Now the simulations are creating simulations. So that's a bit of a problem.

0:10:22 Thomas: Yeah. Yeah. A loop, yeah? Yeah. Yeah.

0:10:34 Bryan: But then it gets to a bigger problem when within the simulation, they start having a conversation, which actually I love how, like, no, they did this in the outermost layer first, having a discussion about whether their world is a simulation or not. And I really love the way it goes. Professor Farnsworth first says, oh, of course we're not in a simulation. Don't be preposterous. And then they go through these phases of Amy pointing out ways that Professor Farnsworth took shortcuts in this simulation, which happened to resemble weird physics things.

0:11:08 Bryan: One of the shortcuts is, well, if Bender moves, the gravitational force from him has to affect every other object in the universe, so I limited that by it only traveling out at a certain speed. You mean like the speed of light?

0:10:48 Thomas: Yeah. Yeah. Yeah, exactly.

0:11:22 Thomas: Right. Yeah. It was it was funny. It was cute. I also wonder, are they trying to it seems like they've tried to make Amy's character better. Like, I feel like maybe that's that's a conscious choice they made because.

0:11:32 Bryan: That's part of it. It actually was in the canon that she's pretty smart. She has like some sort of advanced degree. So they're just using that now.

0:11:36 Thomas: Yeah. Gotcha. Yeah. And so she's the one throwing out all the stuff that's like, so wait, that sounds like quantum physics. That sounds, you know, and I, maybe it's worth worth pausing there. So he says, yeah, you know, one of the ways I'm saving like, you know, computational power here is I have stuff that it doesn't even, you don't even know where the position of it is until anyone looks at it or whatever. And she's like, oh, isn't that like quantum physics? And so I'm wondering just first question, physicists, how much do you feel like that's a thought versus an actual legitimate thought, which is, whoa, this whole quantum physics thing, that that does seem a little weird. Does that fit as a possible explanation or is there some reason why that instantly doesn't fit?

0:11:41 Bryan: Yeah. Yeah, yeah.

0:12:19 Bryan: So I actually have two trains of thoughts on this. One is that if you've ever tried to learn it, the math behind quantum mechanics is horrifically more complicated than the math behind classical mechanics. And so my first thought is that, okay, trying to do all the math for a quantum mechanical representation of things would take a lot more effort than doing a classical classical representation.

The second thought, though, is if you're running the simulation on a quantum computer, there might be a way that you actually could get benefits. Like, if you go back to what we talked about in the previous episode, we talked about there's a minimum amount of energy needed for any sort of irreversible computation.

And one of the big things quantum computers can do is that they limit the number of irreversible computations, and so if things in the simulation follow quantum mechanics and they don't do that irreversable step until someone observes something, so maybe they are saving in all of those computations all along the way, and then they only resolve it when some conscious observer checks on it. Though then you have to have the computer figure out, okay, what is a conscious observer and how do

0:13:39 Thomas: Yeah.

0:13:39 Bryan: You calculate that. And there's, to think of it, quantum mechanics has one very massive problem that regular mechanics doesn't, is that quantum mechanics, if you just follow all of the equations exactly as written and don't do anything special, it naturally leads to infinite parallel universes. Yeah.

0:14:00 Thomas: How would that work with the math? Unless, well, no. Okay. Let's say we're the simulationers. We're the simulators, right? So we Could just say, well, God does play dice with the universe, right? Couldn't we just be like, well, no, we're not making all those those possible other simulation universes, we're just deciding, no, that, uh, what is it, the something interpretation, Copenhagen?

0:14:04 Bryan: Yeah. Yeah.

0:14:22 Bryan: Copenhagen.

0:14:22 Thomas: Something is that, yeah. We're just like, nah. That's not right, we decide as the simulators, that's not the Copenhagen interpretation, meaning, uh, guess, not to go into that, but just that rather every single time there's one of these divergences, it's it's a whole actual another universe.

0:14:23 Bryan: Yeah. Yeah. Just. Yeah. Copenhagen. So the two competing interpretations is multiple universes, where everything splits and keeps splitting ad infinitum. And Copenhagen interpretation, where when things would split, what is randomly chosen to be reality. And what's called a collapse. And so it doesn't expand infinitely. And you just get one, but there's this weird collapse thing, which isn't well explained. But from inside, you wouldn't know the difference. So yeah, the simulators could do this and the simulations wouldn't know the different.

0:14:37 Thomas: Oh. Okay. Gotcha.

0:15:00 Thomas: Gotcha.

0:15:10 Thomas: And it is interesting. There's a sense in which the whole "until something is observed" thing, it doesn't have to be a conscious observer, obviously. And so from simulation standpoint, you can see, well, there that you're simulating that do not need to have firm properties until something else needs to interact with them, right? And so I see the appeal of like, yeah, that seems kind of simulation-y, honestly. Honestly, but I was just wondering if there's any more, you know, reason as to why that would break down. Are

0:15:29 Bryan: Yeah.

0:15:40 Bryan: Now, the other side of this, the speed of light limit, that's actually more straightforward, in that it does save resources, but not all that much. Can kind of think of this like in a two-dimensional universe, like it's easy to draw this, so I'll just try to describe this simply. So let's just say you take out a piece of paper, at the very bottom edge you write out your x -axis this is distance, and the vertical edge is your time axis. And so let's say that right now we're at half the distance on x and and all the way up in time. So let's say that our slope for the speed of light directly connects the bottom left corner to that top center point. And then for the other direction, you go down to the bottom right corner. If you draw this out, every point in space -time that could affect the present is within this triangle we've drawn. This is the light cone thing where you draw out multiple maps of different

0:16:06 Thomas: We Yeah.

0:16:40 Bryan: Size circles and stack them up.

0:16:41 Thomas: Okay. Yeah, we every outhouse. Henhouse. Gotcha, that that makes it fit, right?

0:16:43 Bryan: Yeah, yeah, there we go. Yeah, so the point of this is if you draw this out, you've got a triangle and a piece of paper, so you've cut down the amount of space you need to deal with from the whole paper to just what's inside this triangle, but this triangle still takes up half of the the paper. So in one -dimensional case, you're throwing out half the data you don't need to worry about, which is a definite savings, but it's not a ridiculous savings compared to how hard it is to simulate the entire universe.

0:17:17 Thomas: All Right, well, well, we can get, go back into, so at this realization, I think you were saying you liked the way they did this part. So I want to make sure that you said everything you wanted to there.

0:17:27 Bryan: Like at the end, after Amy points out, Professor Farnsworth turns on a dime and says, well, as I was saying, we probably do live in a simulation.

0:17:35 Thomas: Like, wasn't that exactly the opposite of what you're saying? I changed my mind because of new data. Yeah, that was funny.

0:17:38 Bryan: Yeah, which is Good. That's a good scientist thing to do, is taking in those arguments. So these are some good points about, you of, our universe that make us wonder if we might be in a stimulation. And I also like that they point out things that are true for us as well. And to get us thinking, kind of prompting the question for us, which I'll get into. For now, back in the episode. So the Simulated Universe, the group decides they want to know for sure if they live in a simulation. So they're going to try to trigger something that would be so hard to compute in a simulation that they would see glitches in it, and the idea they come up with is causing the collapse of a magnetic star. Yeah.

0:18:25 Thomas: Yeah, I wait to hear about whether or not that makes sense.

0:18:28 Bryan: So a magnetic star, we talked about this in an episode a while back, uh, a type of neutron star with a very high magnetic field. And this is a part of physics where classical physics does break down. Neutron stars are dense enough that you can't fully describe them with Newtonian gravity. You absolutely need Einsteinian gravity to explain them. And Einsteinian gravity is a lot harder to simulate than Newtonian gravity, um.

0:19:00 Thomas: Question though, hypothetically, if we were the simulators, if we are somehow in charge of this universe, we definitionally would have to have a complete understanding of the physics, right? We, I mean, we've programmed it out unless there's some sort of algorithmic learning thing, you know, or something, but some, something does.

0:19:02 Bryan: Yeah. Yeah. Well, there could be emergent properties that we don't understand, like, we could understand the base level physics but not understand how that leads to human consciousness.

0:19:19 Thomas: Yeah, true. True. Well, that's Just the question I'm going to ask, which is if now you're saying, like, if we're trying to, I assume you're saying if we're trying to simulate stuff on our computers right now, OK, yes, this is a thing that's way harder to do. But what if we discovered some baseline unifying equation or something, something that simplified it out? And it was like, oh, actually, this makes it way easier to simulate this because we have a deeper understanding. Whereas before, we were doing a lot of ad hoc things that are, you computationally difficult and resource intensive because we didn't have a full understanding of physics.

0:19:25 Bryan: Yeah.

0:19:58 Bryan: So that's, uh, possible, and it's not even a full understanding of physics, but there have been cases, uh, we've made great advancements in simulations through mathematical advancements. So I'll give you one example of this, which we use all the time in astrophysics when doing simulations. So let's say we're just doing a simulation of dark matter, which only interacts with itself through gravity. So this is a pretty simple, straightforward simulation to do. Now, when we're doing this, we get to the kind of the problem we talked about before, that every particle is influenced by the gravity of every other particle, though limited by the speed of light. So if you calculate this out, let's say you have this big cube of space you're simulating, and you have billion a particles in there. So when you want to advance time forward one step, you need to take each particle, find the gravity from every other particle. So that's a billion particles times a billion other particles, divided by some savings due to the light cone. So you've got a billion squared, or 10 to the 18, interactions that you have to deal with.

0:21:14 Bryan: So And this is a lot. It's a problem. When we're talking about big computations like this, what we use to describe is what's called big O notation, which tells how fast the computational time scales with the size of the input data. And this is a case that we call order N squared, which we write as big O, parentheses, N squared. N being the number of particles or items or what have you. And this means that if you double the size of the simulation, the computation goes up by the square of that since we're doing N squared, so we go up by 4. You multiply it by 10, it goes up to 100. This makes it prohibitively difficult to do larger simulations, and this is a problem that we have with all of these gravity simulations. Simulations it out, however, that there is a mathematical trick we can use which vastly simplifies and decreases the scaling of this type of simulation. This trick

0:21:14 Thomas: Sounds like a lot.

0:22:19 Thomas: PEMDAS.

0:22:19 Bryan: Is called the no.

0:22:20 Thomas: No.

0:22:21 Bryan: It's called the Fourier transform. Are you familiar with that? Oh, okay.

0:22:25 Thomas: That was in the Good Will Hunting, Law'd Awful Movies, I think. That was related to one of the problems he solves on the board. No joke. Joke.

0:22:33 Bryan: Yeah. Yeah. Okay. Yeah. Yeah, so for anyone who isn't familiar with that, those who aren't patrons, Fourier transform basically takes the distribution of things in space and figures out a different way to map out the data in frequency rather than space. So you can think about this, like, we're recording an audio file right now, and if I go into Audacity, look at what I'm recording, I'll see the wave function of the sound coming out of being recorded at like every millisecond or whatever the sampling is.

Another way you could represent that is the frequencies that sum up to give the resulting function. And I'm pretty sure Audacity has a way to do that. And you can see the frequencies both any given time, or for the full Fourier transform, you can do mathematical tricks and fit the full set of frequencies, which if you add them up, depending on the strength of each frequency, will add together to give the full wave function over time.

So in words, we're basically just shifting things from real space to frequency space. You've - you've got the same amount of information describing it. We're just describing it a different way.

0:23:53 Thomas: It's like describing stuff more. It's like just in bulk versus, you know, every single particle kind of thing. Like, yeah, that makes sense.

0:23:59 Bryan: Or It this way. Like if I were to record in Audacity a constant tone of a single frequency, it'll be recorded as a perfect sine wave. If I do the Fourier transform of that, it will calculate the frequency of that sine wave and give a peak, one point there that'll be up like a value of one at that frequency and zero everywhere else.

0:24:26 Thomas: Right. Gotcha.

0:24:27 Bryan: So Yeah, it works like this. And this turns out to have some very useful mathematical applications, which to work for Newtonian gravity. Not general relativity gravity, Unfortunately, but for Newtonian gravity, it does. It allows us to do a trick which simplifies the calculations from order N squared to order of N times the natural logarithm of N. And the natural algorithm scales much slower than N, so we save a massive amount of time. And, proving this, gets it pretty mathematical. But the trick is that the way gravity works is it's additive. The gravity from any individual particle adds to the gravity from other particles.

And this results in a mathematical operation we call a convolution. We're just adding up the gravity from everything around. And Fourier transform of the data, change it from position space to frequency space, the operations you do on it change, and a convolution becomes simply multiplication. So while a convolution has this N squared scaling, multiplication, you just do it once at each point in space. You're going from N squared to N there.

Now the Fourier transform is difficult to do, but we have ways to do it in N times the logarithm of N time. So instead of having an overall N squared scaling, we do a Fourier transform, which is N times log N. We do a multiplication that's order N, so we're just adding N, but it doesn't add bad scaling to our time. And then we have to do a reverse Fourier transform to get back to position space, which is again N times the log of N. The reverse Fourier Transform is mathematically almost essentially the same as the Fourier Transform, just one of those weird things. So when you you come back, know, there you've are more individual operations going on, but those operations scale much more favorably. So when we get up to these large simulations, we do this trick with Fourier transforms to cut down the amount of computations we have to do, and it's a dramatic reduction.

So instead of doing 10 to the 18 from a billion squared, we would do a billion times the log of a billion, which, I don't know, is around 100 or so, or 10 to 18.

0:24:39 Thomas: Oh, dang.

0:26:57 Bryan: So we cut computations by a factor of a million or so. So

0:26:57 Thomas: That's That's a lot. They could have had that be the plot point where rather than the power, you Know, finding the power from the sewer or whatever that nonsense was.

0:27:06 Bryan: Yeah. Yeah, but remember one important thing I said, or put important things together. One was that neutron stars are dense enough that you can't accurately describe them by Newtonian Gravity. And the other is that this Fourier transform trick only works with Newtonian and gravity and not general relativity. Reason for that is Newtonian gravity is linear, like you double the mass, you double the force, and that's all there is to it. General relativity is more complicated than that because, like, at some point you break space and create black holes, it doesn't quite act linearly, and

0:27:21 Thomas: Right, which would make one wonder or if you were simulating the universe, why would it be that way? I don't know. It seems weird.

0:27:54 Bryan: General relativity is one of those things that, if if you were programming up a simulation, why would you put in something that complicated. And, you know, if you look into the building blocks of physics, you can justify it in some ways. Like if you want to have a speed of light where the speed of light is constant from all perspectives, you have to have special relativity. Go back way to like some of our earliest episodes, we talked about that. And the speed of light is also necessary for magnetism, believe it or not. Like, put that on hold for a future episode. It's

0:28:04 Thomas: Yeah. Mm -hmm. Yeah.

0:28:31 Thomas: Brian, I would believe anything you said about the... It's I mean, it's key to physics, so I guess it's necessary for all of everything, right?

0:28:31 Bryan: Yeah. Yeah. Yeah, with magnetism, that's an important part of the way atoms are built, the way they are. So if you don't have the speed of light, physics looks very different. So yeah, maybe it's a necessary part of that complexity needed for our universe, universe. And then general relativity is necessary to describe what happens when things start accelerating, because if you have special relativity on its own, you can come up with certain paradoxes, like the twin paradox is the classic one, where you have two twins, one goes traveling around at a significant fraction the speed of light, and because of time dilation, when they come back, they're younger than the other twin. But if you flip the perspectives, perspectives around, the twin who is traveling sees themselves as stationary, and the other twin as moving. So how come when they get back together, one of them ends up younger than the other? And you need to go into general relativity to explain this. It could be that this is all necessary to have a sufficiently complex universe. And yeah,

0:29:39 Thomas: Mean I feel like as the simulators, once again, we could have been like, Like, ah, we'll just take that loss. Like, we will take that L. If some twins do that, we would just let them both be older than each other or something. I don't know. We will figure it out. I would be the programmer that gets fired from the project, though. I would say, no, we can't do that. That's impossible.

0:29:50 Bryan: Knows? Yeah. Yeah, what we have in the episode is, I, I kind of see what they're going for, they're going for, um, object in space, which is particularly hard to simulate because you need general relativity for it, and they're gonna make it collapse. So a neutron star collapsing needs to collapse into a black hole, and a magnetic neutron star is going to further complicate things, because with the strong magnetic field lines, those need to somehow collapse as well and form a magnetic layer around the black hole. It does get into very complicated physics.

I think Professor Farnsworth's computer could handle it, and if it's able to just handle all the calculations of the entire universe, which is pretty big, Think then, it could handle, uh, this part. But you, for the sake of the plot, I think it's fine. It is something that is very complex physically, and it's also something that's very interesting to look at, like way they illustrate it in the show, um, shows the magnetic field lines, which, you know, you wouldn't directly see them, but You would see the effects of them, like, which, like, in the real world, you see the affects of our magnetic fields from the northern and southern lights.

0:30:42 Thomas: Yeah. Yeah. Yeah. Yeah. Oh, gotcha. Yeah.

0:31:09 Bryan: So come from charged particles in the atmosphere essentially getting caught in those field lines. And then one law of physics is that accelerating particles, which spiraling particles are, accelerating charged particles will emit light. So particles get trapped, spin around, and emit light, and that causes the the northern and southern lights, when the magnetic conditions are right. And so you'd see this in a magnetic star as well. You would get a lot of light coming out all from the charged particles around it, spinning around and causing a brilliant light show. Got all of that, and then you're collapsing the star, changing it rapidly. You've got to handle the magnetic fields getting close to the event horizon of a black hole. They kind of get frozen into space there. It is horrendously complicated. And yeah, a good way to kind of stretch a simulation to its limit. See how it is at handling all the physics in the extremes.

0:32:07 Thomas: Presumably, this is happening in our simulated universe, but no one is around, so it doesn't take computational power? Or, gotcha.

0:32:15 Bryan: Yeah, that could be. I mean, the entire universe is pretty big. Remember, in astrophysics, time scales are very long. And neutron stars and black holes are pretty rare objects. And the time period for a neutron star collapsing into a black hole, and in particular, Magnetar, a very magnetic neutron star, know, those you know, are going to be pretty rare. And I think they even say in the episode that they might have to wait hundreds of thousands of years for to happen naturally.

0:32:47 Thomas: Uh, I still feel like that wouldn't solve your problem, would it? I mean, if you're simulating a universe, you have to simulate the effects, how that's going to propagate out anyway, don't you? I don't know. Oh,

0:32:57 Bryan: Yeah. Yeah. I part of it is that Professor Farnsworth has just started the simulation and and it hasn't happened yet. They do show in the episodes that there was a brief hiccup when two black holes merge, which is another good example of general relativity causing issues as you absolutely need that for a merger of black holes. Back to the episode. They're having kind of an existential crisis, and then we go back to The Real Universe. Bender is having an existential crisis, too, because somebody reminded him that the simulation running their consciousness is effectively the same as the consciousness in a robot brain, and he decides, well, that means he isn't real, because those people in the simulation obviously aren't real and it's the same code.

0:33:04 Thomas: Okay. Yeah, We get some substrate chauvinism real quick, that's what you call it in the business, yeah, yep.

0:33:50 Bryan: That's it. Only meat brains are real. Metal brains, no, those aren't real. Yeah, is going through a crisis, but eventually he resolves it for him, himself, and decides, okay, well, they're going to do this test, they're going to find out they're not real, but I think I've already kind of gone through this trauma. If I can explain it to them, I can kind of let them down easily, help them through it, help them manage it. So he wants to go into the simulation.

0:33:55 Thomas: Yeah, This part's kind of weird, but we'll go with with it

0:34:28 Bryan: Yeah, yeah, we'll go with that. Professor Farnsworth says, okay, we can put you in, but then what happens to the bender inside? Well, that bender gets pushed into the sub -simulation. What happens it to that bender. Well, to the sub -sub simulation. It's benders all the way down.

0:34:29 Thomas: Yeah not sure why that would be. Yeah, so it just, it shoves it's that. Oh, what's the infinite hotel thing, what is that one? Remember that? Yeah.

0:34:45 Bryan: Yeah. The Hilbert Hotel, I think that's called.

0:34:51 Thomas: Yeah, It's like that with benders, though, so it's, if, the from the upper, you know, right?

0:34:55 Bryan: If There's a hotel with an infinite number of rooms, and each room has a bender occupying it. A new bender comes along and wants a room at the hotel. How do they make room? Well, you go to the bender in room one, tell you, okay, you move to room two. Then the one in room 2 moves to room 3, room 3 to 4, and it's an infinite number of rooms, so you're never going to run out of rooms Yeah, so you just move them all down one level, so as long as you accept that there are infinite simulations, this works, though. Accepting that there are infinite simulations is a bit of a trick, but we'll just go with it.

0:35:12 Thomas: There You go So you're all set.

0:35:34 Thomas: There's also, I shot, that's really, that's good, that I wanted to point out, which is before the first or second commercial break, when, when they have that realization where they're like, but what if we're in a simulation, you know, kind of thing, um, the lower, the simulation, Futurama people realize that, and it goes, "Ah," and then it pans out to the real, you, reality, and they go, "Ah," and then it does like a fake pan out again a little bit, and then it goes to commercial break. I thought that was really clever, yeah.

0:36:00 Bryan: Oh, yeah, that was good. Yeah, over here, they have it on Disney+, which was without commercials. And this might be the one time in history that that's not a good thing.

0:36:05 Thomas: Oh, I I mean, I, yeah, it was on Hulu, so it wasn't an actual commercial, but you could just tell it was like that, you of, they do three acts usually to these cartoons, you know, so, it, it's, it wasn't, it doesn't, it didn't, you don't, so I, I don't know. It was, uh, it it was before a fade out kind of thing, right?

0:36:10 Bryan: Yeah. Oh, okay. Yeah. Oh, Yeah. Yeah. Yeah. Okay. Yeah. That was a good one. Yeah. You know, there was actually another thing I was using. I was playing a video game which tackled this. It was called AI: The Nirvana Initiative. And I think the video game actually has a neat twist to this as, you know, I'm playing along and I come across this one character who has this weird philosophy. She is convinced that the world she lives in is a simulation. So I'm sitting there playing this game on my Switch, looking down at her on screen. I said, well, you're not wrong because a game, even more so than a TV show really is a simulation.

0:36:53 Thomas: Mm-hmm.

0:36:54 Bryan: The TV show is, you know, kind of a recording, but it, it adds that philosophical layer to it. It kind of reminds us that, yeah, we know they in a not real reality. And then it kindof raises the question to us, how do we know that we're in reality? That's something we'll get to in a bit. So, The episode, Bender wants to go in and, p- does pushes the Benders down, our Bender is gone, and know, you as they're talking about it, they actually talk it through, decide, you know what, I think I'll be okay with it, I just want to know, and Bender decides, yeah, know what, they're okay, I'm not gonna break it to them, I'm just gonna go along and

0:37:14 Thomas: Sure.

0:37:36 Thomas: Right. Bender, so Bender's in the simulated universe, and for some reason, again, he has to go down there to tell them, that seems a little contrived, but but, he, yeah Exactly, um, but he sees that, you know, the connections of the characters and how that's all that matters to them. And they don't want to know whether or not they're in a simulation. They don't care. But it is glitching this whole time. And I guess I wasn't sure if like so are the characters not perceiving that you think? I'm not sure.

0:37:42 Bryan: Yeah. Yeah, why you just program in a message from God?

0:38:03 Bryan: No, they are. I think there was a line that somebody, when they collapse the Mangatar, somebody does say so pixely. I think it might have been Zoidberg. So they do see it, and they've just decided they're okay with it.

0:38:10 Thomas: Oh, yeah. Gotcha. OK.

0:38:16 Bryan: Yeah. So then back in the main lair, Bender suddenly comes back to life for no reason explained and says, is, oh, wait, I figured out how to keep the simulation from crashing because it's going to use up so many more calculations trying, struggling to simulate the collapse of the magnetar.

0:38:36 Thomas: Yeah. And they didn't want them to die in there. They were trying to save that rescue this the simulants.

0:38:38 Bryan: Yeah. Yeah, which is kind of a good point that you've got consciousnesses. It would be essentially genocide to turn off the simulation. You would be killing everyone for the sin of living in the wrong universe. So you've got to keep it going now you've started it, which I think personally is a good moral argument against ever doing the simulation in the first place.

0:38:55 Thomas: Yeah. Yeah. Totally.

0:39:04 Bryan: They've keep it go. And Bender says, okay, I know how to do it. You just have to slow down the simulation. Them, time will appear as normal. But for us, it'll be be going slow enough that we can keep things from crashing here we can get through it.

0:39:19 Thomas: Yeah, You're adding, uh, uh, essentially you're Buffering, yeah.

0:39:19 Bryan: Yeah. Yeah. Something like that, yeah, um, turns out to work and you go back down one layer and you see everybody in the simulation watching the star and things slow down as they're watching it and things glitch out and nobody has an existential crisis, they just think it's really cool to see this simulation kind of, but not entirely, break down. That's basically where the episode ends. And it's not where the interesting stuff ends, as it really gives us a Lot of Interesting Things to Talk About, because this is basically putting some very strong questions in front of us and asking us to grapple with them. And I think the first one, and this is kind of a softball, is why did Bender reactivate in the outermost layer we see?

0:39:55 Thomas: Yeah.

0:40:11 Thomas: I do like that they don't try to explain it at all, and I think that we're to assume it's some sort of loop, you know, where you've you've simulated all the way down to where it loops in on itself, and hotel, you've kind of come back to room room zero, you know, or whatever, and and that's how I interpreted it. But what did you Think.

0:40:29 Bryan: Yeah, yeah. I interpreted that it wasn't a loop. It does go down infinitely. We're just not watching the top layer of the simulation. And top layer or even just a higher layer, they figured this out and they see the simulation below Lo is having this problem. And if one simulation has this problem, it's going to propagate upwards because they're to simulate the simulation simulating and crashing. So they have to send Bender down to prevent this. So that's why we get the new Bender.

0:41:01 Thomas: It's kind of the same thing really, I guess, I don't know, I'm not sure.

0:41:05 Bryan: Yeah. Well, there is is one other problem I spotted, this is in Futurama, the universe has a finite age to it, established way back and actually reaffirmed in a more recent episode in this season, that the universe has an finite lifespan to it and loops over itself. It basically starts over and does exactly the same thing. This means that if you slow down one simulation, there's going to to be less time in that simulation until our universe ends. So if that simulation goes 10 times slower, they're going to have one-tenth the time. And then the simulation inside that is going to have a one-hundredth the time, and in that, one-thousandth. And as you go all the way down, all way down, eventually you're going have simulations that barely have any time at all. So it looks like it solves the problem, but upon thinking about it, those simulations are going to end from the universe ending, and they'll end after, you know, a very short amount of time, it

0:42:07 Thomas: Okay, so you're saying that doesn't solve the problem, I mean, it solves the problem to a certain extent but not If we're Assuming An Infinite regress all the way down, I

0:42:13 Bryan: Helps. I Mean. Yeah. And to the extent that's kind of like what you medical science is, you're not preventing someone from dying, you're giving them more time. And so instead of the simulation crashing immediately, you slow it down and you can give them more time up to until when our universe ends. So it's certainly a good thing. It's, yeah,

0:42:38 Thomas: Think. Morally that's cool though, like, like any moral rule I think is going to be, well, like, like if the universe ends, you're good. You've done it. Like you got to keep these people alive up and until the entire universe ends. And so for each simulation, they're good, they're holding their end of the bargain.

0:42:45 Bryan: Yeah. Yeah, they're best and doing what they get. I think, create a simulation in the first place is the morally best option. But once you do create one, you can't turn it off or you're doing a genocide. Want to ask you a question, Thomas. Do you think we live in a simulation?

0:43:01 Thomas: Sure. Yeah.

0:43:12 Thomas: Think it's impossible to know. And I think that my main complaint about all of this has to do with, it you know, actually, I think I covered at one point, what's the paper by the guy? It's not Parfit, is it?

0:43:25 Bryan: So let's see. Was it Nick Bostrom?

0:43:27 Thomas: Bostrom, that's it. That one is interesting. I talked to somebody about that way back in the day, philosopher, I think.

0:43:27 Bryan: Yeah. Maybe Philosophers in Space did an episode on it.

0:43:35 Thomas: I did this before that, I believe. And it was, I tweeted on this show, if I'm remembering correctly. And I find it interesting. It's so, did we, were you planning on talking about that paper at all?

0:43:37 Bryan: Okay. Okay.

0:43:46 Bryan: That's actually what I was going to get into next.

0:43:48 Thomas: Perfect. Well, I would be from memory, so you can feel free to fill it in better, but it's it's essentially, uh, um, speaking Bostrom believes to have proven that we either live in a simulation or we don't get to the point where we can make simulations. And it's because if simulations are abundant someday in the future, then that that would mean like there's going to be far more simulated universes than are not simulated universes. By, and you know, potentially astronomical numbers there, if you take this example, this Futurama for an example, like all the simulated universes are astronomical compared to the apparent one real reality, if we're to believe that, right?

So the odds that any one person, if you're just thinking about out yourself here, you're like, what are the odds I live in the one real universe? Like the odds are astronomically low that you would live in that universe. And so it's, thinks by my memory again, it's been a few years, but it's, we can say for pretty much certain that either you live in a simulation, we live in simulation, or for some reason societies never get to that point where they can simulate that many universes for one reason or another, could be computational limits, could be, it doesn't really matter, but like, for what, what, that's like kind of the dilemma. It's like one of those two things has to be true, according to this paper. Did

0:43:52 Bryan: Yeah. Yeah yeah yeah.

0:45:05 Thomas: I do an okay job, or do you, is there anything other wrinkles to it? Oh, cool, yeah.

0:45:06 Bryan: Oh, that was almost perfect, because most of the time people talk about that argument, they forget half of what he said. They focus on the yeah in the future people are going to make all of these simulations, and because they're going to be so many, we live in it. But they forget that he said, okay, it's also possible that for some reason they'll never make those. Maybe it's impossible. And if that's the case, then yeah, we won't live in a simulation.

0:45:14 Thomas: And they just say we live in a simulation, yeah. Yeah yeah Maybe it's impossible, maybe, you know, it's not it's it's Fermi equation, uh, no, Drake equation. Yeah.

0:45:32 Bryan: Oh Equation, that's it, yeah. Mind was going for me, which is for Fermi. Fermi paradox is probably related, but Fermi estimation, in previous episode mistakenly called it Newtonian estimation, but yeah, it's, yeah, it, Fermi estimated is what the Drake equation does, and then the Drake Equation is just a specific example of that.

0:45:42 Thomas: It's the Fermi paradox, or oh, right. Yeah. Gotcha. Then was the Furby Paradox, why haven't we met these aliens? Okay. So we got there, everybody. And so if remember correctly, that could be one of the possibilities too. If there's some reason, like some inherent reason that if societies or alien races or whatever planets get like necessarily the amount of technology it takes to get to that point is just so, you of it are such that like you end up killing yourself of climate change or something related or can't manage all that. There could be systemic kind of reasons why that's just a necessary thing. And so that could be a reason that there won't be simulations. Or it could be like you you get that advanced, then you come up with rules that you shouldn't make them. There's lots of possibilities of why you you know, like why that side of the equation in the Boston paper could obtain, as they say. But like, yeah, I do also, I found the same thing that everybody kind of forgets that part of the paper and they're just like, we live in a simulation.

0:46:01 Bryan: Exactly. Yeah So this presents an argument that we can actually tackle both mathematically and scientifically. It's a tricky one, but that's kind of the point I wanted to get here. Is let's take this seriously and work through how much sense this makes. And

0:47:15 Thomas: Right, folks, that's where we're going to leave it for part 1. A little bit of a cliffhanger there where we get to hear Brian take on the simulation hypothesis in part 2. Got a unique take on it as well. It involves, as a teaser, something called the Sleeping Beauty problem. So we get to walk through another fun philosophical thought experiment. I love philosophical, slavical thought experiments. I always have. I love them so much. And so I'm a sucker for this kind of thing.

Can't wait to get to part two. I was actually going to tidy up my summary of the Bostrom paper in of the post, postscript here. But as I listened back to it, it was actually pretty close, and that was from memory from years ago. But just because we never read it like word for word, I want to make sure since we're going to be going through this argument and critiquing it possibly, seeing where we land, I want make sure we're representing.

So I do want to read Bostrom's conclusion. And the one thing I did kind of blend together was two of the three branches of the trilemma, as they call it. Even though I think I pretty much said them all, but just to state it accurately for the record here, so that we're all on the same page going forward to future episodes. So the conclusion, I'll just read the conclusory paragraph of Bostrom's simulation argument. Quote: a technologically mature post have enormous computing power.

On this empirical fact, the simulation argument shows that at least one of the following propositions is true. One, the fraction of human level civilizations that reach a post -human stage is very close to zero. Out. So that's the kind of the equivalent of there's some reason that once any civilization becomes as advanced as you would need to be, or somewhere before that necessarily, tend to die off or have some catastrophic thing. You know, think climate change, or killing each other with war. You know?

So that is one possibility. For whatever reason, don't reach a, quote, post -human stage. To the quote. Two, the fraction of post - human civilizations that are interested in running ancestor simulations is very close to zero. And three, the fraction of all people with our kind of experiences that are living in a simulation is very close to one. If 1 is true, then reaching post -humanity.

If 2 is true, then there must be a strong convergence among the courses of advanced civilizations so that virtually none contains any relatively wealthy individuals who desire to run ancestor simulations and are free to do so. 3 is true, than we almost certainly live in a simulation. In the dark forest of our current ignorance, it seems sensible to apportion one's credence roughly evenly between one, two, and three. Unless we are now living in a simulation, our descendants will almost certainly never run an ancestor simulation, quote.

You can see why this is kind of alarming. It seems to, in a weird way, bootstrap itself into the way the universe would work with seemingly not much information. Almost, that's why it seems a little tough to swallow. It's like, well, how, how could we be doing so little work argument, like argumentatively, or scientifically, or philosophically, and yet we get a pretty firm conclusion that like one of those things has to be true, you know?

So if we go on to create simulations, he says we are just about certainly in a simulation, which I think seems pretty strong, to put it mildly. So, are there any problems with that argument? There are, what are they? Does that have to do with Sleeping Beauty? Well, find that out in next episode. I want to thank you so much for listening. Thanks, Brian Gillis, as always, for coming on the show. And we'll see you next time.