Live data from Hacker News

On whether we're living in a simulation

scottaaronson.blog

71–80 of 206 posts

Re: On whether we're living in a simulation

#71
post #49

Earlier quoted context omitted.

"The whole simulation hypothesis seems like a new way to introduce theological concepts with modern tech. Cannot be disproved, cannot be verified." I have bad news for you. There is no proof in Science. :-( The problem is, that if it is possible to simulate us, then the likelihood that we are simulated is much higher, than that we are not. This is a problem! And yes, can be tested. "Stacking Problem". The simulation…

Erm yes, there is proof in science. Despite the fact, that nothing is 100% certain, you come up with a hypothesis and test them and verify or dismiss them. This is fundamentally different to religion, where believe and not proof is required. And if you find, there are convincing hypothesis for the simulation theory, that can be scientifically tested, please go test them. If you are succesful, you will convince people…

A lot of cosmology rests on far less stable ground than most might realize. So for instance the big bang was hypothesized by observing than everything is expanding. So if you roll back the tape then it seems obvious that everything must have been closer. Roll it way back and you get the basic foundation of the idea.

So one natural consequence of this explosion would be the cosmic microwave background radiation, the radiant heat left after the explosion. And we do observe that! Kind of... The thing is, it's nothing like what we expected. When we look at the CMB it's extremely homogeneous, everywhere. The reason this is a problem is because areas that are not causally connected (could not have reached each other at the speed of light, since the beginning of the universe) actually seem to be causally connected. We would expect to see a much more heterogeneous universe after a big bang.

So we resolved this by inventing cosmic inflation. [1] The idea is that somehow, shortly after the big bang, everything just sped up dramatically - well past the speed of light. And then it slowed down. And then started accelerating again in a fashion similar to what we currently observe. If you assume this, then we can get the big bang to match what we observe. But there's no causal mechanism or anything of the sort proposed. It's literally just retrofitting a hypothesis to avoid having it falsified, an ad hoc hypothesis. [2]

Inflation does fit what we see quite well, but it's untestable, unfalsifiable (especially given the ad-hoc nature of its own existence), and the predictions its offered to date were not particularly unexpected, given the homogeneity of the CMB.

[1] - https://en.wikipedia.org/wiki/Inflation_(cosmology)

[2] - https://en.wikipedia.org/wiki/Ad_hoc_hypothesis

Re: On whether we're living in a simulation

#72
post #7

"If they say, maybe you should live your life differently, just from knowing that you might be in a simulation, I respond: I can’t quite put my finger on it, but I have a vague feeling that this discussion predates the 80 or so years we’ve had digital computers! Why not just join the theologians in that earlier discussion, rather than pretending that this is something distinctive about computers? Is it relevantly dif…

"The whole simulation hypothesis seems like a new way to introduce theological concepts with modern tech. Cannot be disproved, cannot be verified." I have bad news for you. There is no proof in Science. :-( The problem is, that if it is possible to simulate us, then the likelihood that we are simulated is much higher, than that we are not. This is a problem! And yes, can be tested. "Stacking Problem". The simulation…

Well, yes, but the "real world" might have infinite dimensions, and the only reason we don't is because we're near the bottom of an infinitely high stack.

The problem with trying to make assumptions like these is that without "outside knowledge" we're stuck trying to postulate limitations based on a world that might well be a result of the "outside world" taking liberties for the sake of computability of their simulation.

The original simulation hypothesis took simplifying steps for the sake of "selling it" that at the same time makes it easier to attack. E.g., the "ancestor simulations" presumes some degree of similarity between those running a simulation and the subjects of the simulation, but there's no reason for us to presume that simulations of our own past would dominate sufficiently if we gain the ability to simulate worlds that we can assume that anyone running "our" simulation if we are in one would represent our "descendants".

Beyond not knowing if "our" physics reflects anything resembling the complexity of a containing simulation, we also need to presume a whole lot of things about "our" simulation to assume that it is expensive/complex to run. A small selection of such assumptions:

1. That the simulation actually simulates the whole of "our" universe.

2. Or even a significant part of it.

3. And that it simulates a long time period

4. Or even more than a single moment over and over.

5. That the world outside our individual mind exists

6. That there aren't all kinds of other crude shortcuts (chunks despawning and serializing differences from base simulation whenever nobody "interesting" observes it, etc.)

7. That we can know there aren't crude shortcuts because we're not seeing any. See 3, 4, 5, plus maybe the only reason "our" simulation is running is because nobody has spotted any shortcuts and triggered kill switches yet ("oh no, another experiment contaminated; time for a reboot")

8. As a variant on 5, that other people, or any significant number of people, exist, and make pruning the simulation computational budget hard.

You can come up with an endless number of these. We can perhaps start to make some very vague guesses about some very generic qualities of a simulation, but the'd largely be in relation to qualities of a hypothetical "outside" that we don't know the qualities off. E.g. it may be reasonable to assume that our physics would either be no more complex than theirs or that our simulation would be strictly limited in scope, be it time, space, fidelity. But that wouldn't tell us much.

From a scientific point of view, this makes the entire hypothesis rather pointless, unless we run head-first into extremely clear signs we are in one. But from a philosophical point of view it's what makes it fun to think about consequences.

Re: On whether we're living in a simulation

#73
post #57

Earlier quoted context omitted.

It's still interesting to think about though. Specifically, I find it interesting that the concept of a "god" is actually realistic in a very tangible way.

As an atheist, the simulation argument, or variants about it, is the closest I've come to considering the belief in a "god". I still stick by my atheism on the basis that I don't feel someone simulating us would really meet what we think of as a deity, and in that sense it has even made me see a "god" as less likely, because I'd consider it more likely that if there were to be anything outside our universe it's just…

Which sounds as unlikely as Jesus dying for my sins on the cross, a very strange thing that falls down upon scrutiny, which also happens to be something I have faith in!

To someone who engages with theology, beyond a happy little bedtime story to make me feel comfortable with my own demise, the Simulationists are in the same ballpark, but without the test of time and some important calls to an almost unattainable set of criteria to live by, best expressed in the Beatitudes.

If pressed I am probably more engaged by an agnostic understanding of the Abrahamic religions, but I can’t help to feel the overwhelming truth, love and warmth of all of existence radiate from the verses of the Bible, Old and New.

My Jesuit theological trainings have led me to appreciate all other religions, to see the same love and energy in the multitudes of practice and expression, like the light in the center of a lamp, throwing subtlety different shadows on the walls from the inconsistencies in the making of the glass panes by human hands.

Perhaps with time the Simulationists will end up with their own window into the glory of our existence! I will welcome such a transformation with open arms.

I want to follow in Christ’s footsteps because it would make me a better person, no matter the fact that the struggle will never end. With God’s grace I am at least given the chance to try, fail and be forgiven for the limits of my human nature.

The emphasis should be on the practice of humility, compassion, understanding, and forgiveness. The particular expression is in the realm of mere mortals and is as beautiful as any other artistic creation.

I expect downvotes, but that’s OK. Father, forgive them; they know not what they do. (Luke 23:34)

;)

Re: On whether we're living in a simulation

#74
post #49

Earlier quoted context omitted.

Erm yes, there is proof in science. Despite the fact, that nothing is 100% certain, you come up with a hypothesis and test them and verify or dismiss them. This is fundamentally different to religion, where believe and not proof is required. And if you find, there are convincing hypothesis for the simulation theory, that can be scientifically tested, please go test them. If you are succesful, you will convince people…

"Erm yes, there is proof in science." Well, then you know more than me. And I have a STEM PhD. " But I am open to be convinced with good arguments." I told you. Stacking argument. There is a paper about it, I can't find it right now. And that something is not testable right now (Think Einstein Relativity that was for some years only a theory) or String Theory does not mean that it can't be tested in the future. The p…

Are you seriously arguing, that there is no such a thing as scientific proof?

I think I stated clearly the philosophical limitations of that "Despite the fact, that nothing is 100% certain" but that still makes proofs a scientific thing.

"Einstein Relativity that was for some years only a theory"

And Einsteins theory was indeed controversial at first, but it was a solid piece of work, that indeed turned out to be true by experiment. But people were convinced before that, because it explained things in a satisfying way.

And string theory is indeed just a theory. But one that aims to explain something. (I do not find it convincing btw. but it is a somewhat working model)

But the stacking argument ... what exactly is it explaining? And when can it be tested?

Re: On whether we're living in a simulation

#75
post #22

Earlier quoted context omitted.

That variant is simply Pascal's Wager. It's more interesting to ask whether, of we are in a simulation, it is possible to prove and/or find "escapes". All our experience with computational sandboxes so far is that preventing information leaks or escapes is extremely hard. Of course, we still have the problem that if we fi d nothing we don't know what that means. From my perspective, I'm on the side that it's a fun ph…

There is no definition of what "simulation" actually means let alone the definition of escape. You're just starting with something you know, ie simulating something on a computer or other device and assuming there is something comprehensible outside it, just like how our reality exists outside a computer simulation. But consider the reality that software exists in and what use our reality is to it. It cannot exist in…

Of course there are plenty of definitions of what a simulation would mean, and I can give you a clear definition of what I meant by "escape".

In this context, I'm using simulation to mean any computational process that gives rise to my subjective experience if that subjective experience does not reflect the world in which that computational process runs. A computer would seem more likely, in a very broad sense, but computation is everywhere.

An "escape" in the sense I used it would be any means by which it were possible to find a flaw in the simulation that would allow us to either exfiltrate information and/or transfer computation out of it.

And, no I'm not saying it's possible. I'm saying it's interesting to think about the possibility.

> You're just starting with something you know, ie simulating something on a computer or other device and assuming there is something comprehensible outside it, just like how our reality exists outside a computer simulation.

No, I'm starting with the question of whether if we are in a simulation, there is something outside that we could find a way to interact with, and if so what the potential consequences would be.

Yes, this is all wild speculation - that is the entire point.

> In any possible "simulation" we live in, "outside" isn't going to entail any elements of what you perceive inside your (created) reality, just like software, other than with convenient assumptions to prop up the idea.

In an infinite set of possible simulations, the outside would entail elements of what we perceive inside our reality. In another infinite set, the outside would entail no elements. We don't have any knowledge that lets us presume to know what such an "outside" would be like, nor which would be more likely, because we don't know who would be running such simulations, or why.

> It may be a fun "mind wank" but don't take it seriously, because it doesn't make any sense.

It doesn't make sense to let it affect us without evidence, but the question matters because understanding it is to understand the limitations of our universe. E.g. anything that might lead us to believe it is less likely we are in a simulation would imply the odds are higher that we will face substantial limitations on computation and/or filters in the future (because if we find ourself not in a simulation, while we might just be first by chance, it might also imply that simulations will be rare or never happen). Whether we can find such indicators is another question.

Conversely, if we are in a simulation, then it raises interesting questions about whether it is possible to learn anything about the potential from that from probing the limitations of "our" physics etc., and whether there are any interesting testable hypotheses there that might tell us something about the odds.

It may never get us anywhere, but so what?

Re: On whether we're living in a simulation

#76
post #57

Earlier quoted context omitted.

As an atheist, the simulation argument, or variants about it, is the closest I've come to considering the belief in a "god". I still stick by my atheism on the basis that I don't feel someone simulating us would really meet what we think of as a deity, and in that sense it has even made me see a "god" as less likely, because I'd consider it more likely that if there were to be anything outside our universe it's just…

Which sounds as unlikely as Jesus dying for my sins on the cross, a very strange thing that falls down upon scrutiny, which also happens to be something I have faith in! To someone who engages with theology, beyond a happy little bedtime story to make me feel comfortable with my own demise, the Simulationists are in the same ballpark, but without the test of time and some important calls to an almost unattainable set…

A "Simulationist" who believes we are certain to be in a simulation despite no evidence is just holding yet another religious belief.

I see it as a possibility, but a possibility without evidence remains an interesting, untested hypothesis that might be grounds for speculation, not belief.

Re: On whether we're living in a simulation

#77
post #57

Earlier quoted context omitted.

As an atheist, the simulation argument, or variants about it, is the closest I've come to considering the belief in a "god". I still stick by my atheism on the basis that I don't feel someone simulating us would really meet what we think of as a deity, and in that sense it has even made me see a "god" as less likely, because I'd consider it more likely that if there were to be anything outside our universe it's just…

Which sounds as unlikely as Jesus dying for my sins on the cross, a very strange thing that falls down upon scrutiny, which also happens to be something I have faith in! To someone who engages with theology, beyond a happy little bedtime story to make me feel comfortable with my own demise, the Simulationists are in the same ballpark, but without the test of time and some important calls to an almost unattainable set…

> My Jesuit theological trainings have led me to appreciate all other religions,

I don’t know much about Jesuits. Is this common in their beliefs or are you just mentioning that you yourself came to these conclusions? Just curious for my own sake.

Re: On whether we're living in a simulation

#78
post #37

The OG Nick Bostrom argument [1] makes an argument for simulation theory with some reasonably simple math you can see in the paper with only a few variables: - `f_p` - Fraction of all human-level technological civilizations that survive to reach a posthuman stage - `f_I` - Fraction of posthuman civilizations that are interested in running ancestor-simulations - `N_I` - Average number of ancestor-simulations run by a…

One of Aaronson's arguments in the article boils down to the idea that running a full universe simulation (without cheating) on a universe with the same physics as ours may just not be possible computationally; it seems physically plausible that you need a universe to compute a universe.

If that's true, then the simulators would need to be running in a different kind of universe than ours... in which case "ancestor simulation" doesn't really make sense.

Re: On whether we're living in a simulation

#79
post #76

Earlier quoted context omitted.

Which sounds as unlikely as Jesus dying for my sins on the cross, a very strange thing that falls down upon scrutiny, which also happens to be something I have faith in! To someone who engages with theology, beyond a happy little bedtime story to make me feel comfortable with my own demise, the Simulationists are in the same ballpark, but without the test of time and some important calls to an almost unattainable set…

A "Simulationist" who believes we are certain to be in a simulation despite no evidence is just holding yet another religious belief. I see it as a possibility, but a possibility without evidence remains an interesting, untested hypothesis that might be grounds for speculation, not belief.

What evidence do you have that there is no evidence though? Or does a different standard apply to certain claims, like cultural axioms?

Re: On whether we're living in a simulation

#80

Earlier quoted context omitted.

Which sounds as unlikely as Jesus dying for my sins on the cross, a very strange thing that falls down upon scrutiny, which also happens to be something I have faith in! To someone who engages with theology, beyond a happy little bedtime story to make me feel comfortable with my own demise, the Simulationists are in the same ballpark, but without the test of time and some important calls to an almost unattainable set…

> My Jesuit theological trainings have led me to appreciate all other religions, I don’t know much about Jesuits. Is this common in their beliefs or are you just mentioning that you yourself came to these conclusions? Just curious for my own sake.

It is common enough that it came out of the mouth of a priest during a theology lecture!

Edit: It is still astonishing to me that they let these hippies into the papacy!

Post reply on HN