0 comments
Comments sorted by top scores.
comment by JBlack · 2025-01-07T00:51:06.994Z · LW(p) · GW(p)
You make the assumption that half of all simulated observers are distinctively unique in an objectively measurable property within simulated worlds having on the order of billions of entities in the same class. Presumably you also mean a property that requires very few bits to specify - such as, if you asked a bunch of people for their lists of such properties that someone could be "most extreme" in, and entropy-code the results, then the property in question would be in the list and correspond to very few bits (say, 5 or fewer).
That seems like a massive overestimate, and is responsible for essentially all of your posterior probability ratio.
I give this hypothesis very much lower weight.
Replies from: AynonymousPrsn123↑ comment by AynonymousPrsn123 · 2025-01-07T01:21:35.457Z · LW(p) · GW(p)
That makes sense. But to be clear, it makes intuitive sense to me that the simulators would want to make their observers so 'lucky' as I am, so I assigned 0.5 probability to this hypothesis. Now I realize this is not the same as Pr(I'm distinct | I'm in a simulation) since there's some weird anthropic reasoning going on since only one side of this probability has billions of observers. But what would be the correct way of approaching this problem? Should I have divided 0.5 by 8 billion? That seems too much. What is the correct mathematical approach?
Replies from: anon-user, JBlack↑ comment by Anon User (anon-user) · 2025-01-07T03:40:08.844Z · LW(p) · GW(p)
Think MMORPGs - what are the chances of simulation being like that vs a simulation with just a few special beings, and the rest NPCs?. Even if you say it's 50/50, then given that MMORPG-style simulations have billions of observes and "observers are special" ones only have a few, then an overwhelming majority of simulates observers are actually not that special in their simulations.
Replies from: AynonymousPrsn123↑ comment by AynonymousPrsn123 · 2025-01-07T03:52:39.927Z · LW(p) · GW(p)
Thank you Anon User. I thought a little more about the question and I now think it's basically the Presumptuous Philosopher problem in disguise. Consider the following two theories that are equally likely:
T1 : I'm the only real observer
T2: I'm not the only real observer
For SIA, the ratio is 1:(8 billion / 10,000)=800,000, so indeed, as you said above, most copies of myself are not simulated.
For the SSA, the ratio is instead 10,000:1, so in most universes in the "multiverse of possibilities", I am the only real observer.
So it's just a typical Presumptuous Philosopher problem. Does this sound right to you?
↑ comment by JBlack · 2025-01-07T01:39:59.027Z · LW(p) · GW(p)
There is no correct mathematical treatment, since this is a disagreement about models of reality. Your prior could be correct if reality is one way, though I think it's very unlikely.
I will point out though that for your reasoning to be correct, you must literally have Main Character Syndrome, believing that the vast majority of other apparently conscious humans in such worlds as ours are actually NPCs with no consciousness.
I'm not sure why you think that simulators will be sparse with conscious entities. If consciousness is possible at all for simulated beings, it seems likely that it's not some "special sauce" that they can apply separately to some entities and not to otherwise identical entities, but a property of the structure of the entities themselves. So in my view, an exceptionally tall human won't be given "special sauce" to make them An Observer, but all sufficiently non-brain-damaged simulated humans will be observers (or none of them).
It might be different if the medically and behaviourally similar (within simulation) "extremest" and "other" humans are not actually structurally similar (in the system underlying the simulation), but are actually very different types of entities that are just designed to appear almost identical from examination within the simulation. There may well be such types of simulations, but that seems like a highly complex additional hypothesis, not the default.
Replies from: AynonymousPrsn123↑ comment by AynonymousPrsn123 · 2025-01-07T02:10:44.738Z · LW(p) · GW(p)
I suspect it's quite possible to give a mathematical treatment for this question, I just don't know what that treatment is. I suspect it has to do with anthropics. Can't anthropics deal with different potential models of reality?
The second part of your answer isn't convincing to me, because I feel like it assumes we can understand the simulators and their motivations, when in reality we cannot (these may not be the future-human simulators philosophers typically think about, mind you, they could be so radically different that ordinary reasoning about their world doesn't apply). But anyway, this latter part of your argument, even if valid, only effects the quantitative part of the initial estimates, not the qualitative part, so I'm not particularly concerned with it.
Replies from: JBlack↑ comment by JBlack · 2025-01-07T02:37:17.884Z · LW(p) · GW(p)
The largest part of my second part is "If consciousness is possible at all for simulated beings, it seems likely that it's not some "special sauce" that they can apply separately to some entities and not to otherwise identical entities, but a property of the structure of the entities themselves." This mostly isn't about simulators and their motivations, but about the nature of consciousness in simulated entities in general.
On the other hand your argument is about simulators and their motivations, in that you believe they largely both can and will apply "special sauce" to simulated entities that are the most extreme in some human-obvious way and almost never to the others.
I don't think we have any qualitative disagreements, just about what fraction of classes of simulated entities may or may not have consciousness.
Replies from: AynonymousPrsn123↑ comment by AynonymousPrsn123 · 2025-01-07T03:06:54.757Z · LW(p) · GW(p)
Yes okay fair enough. I'm not certain about your claim in quotes, but neither am I certain about my claim which you phrased well in your second paragraph. You have definitely answered this better than anyone else here.
But still, I feel like this problem is somehow similar to the Presumtuous Philosopher problem, and so there should be some anthropic reasoning to deduce which universe I'm likely in / how exactly to update my understanding.
comment by CstineSublime · 2025-01-06T22:25:21.701Z · LW(p) · GW(p)
I'm afraid I don't understand a lot of your assumptions. For example, why you think you being an example of any given superlative is somehow a falsifying observation of the reality - especially if other people/objects don't exist in uniform distributions. So it's not like a video game where every other NPC exactly 10 HP, but through use of cheat-code you've got 1000. And even so, that data from within the 'simulation' as you call it is not proof of something 'without'. I think the only evidence of that would be if you find yourself in a situation like Daffy Duck, the walls of reality closing in on you - meeting your maker directly.
I also wonder, how much Kant or Plato have you read and did you do any research, even on SEP before you asked? I feel like anyone who has questions about the 'simulation' would be best served reading the philosophers who have written eloquently on the matter of how we come to represent the world and really formed the concepts and language we use.
Or you could read (Tractatus) Wittgenstein and dismiss all metaphysics all together as nonsense - literally: that which cannot be sensed and therefore musn't be spoken about.
↑ comment by AynonymousPrsn123 · 2025-01-06T23:49:51.648Z · LW(p) · GW(p)
My argument didn't even make those assumptions. Nothing in my argument "falsified" reality, nor did I "prove" the existence of something outside my immediate senses. It was merely a probabilistic, anthropic argument. Are you familiar with anthropics? I want to hear from someone who knows anthropics well.
Indeed, your video game scenario is not even really qualitatively different from my own situation. Because if I were born with 1000 HP, you could still argue "data from within the 'simulation'...is not proof of something 'without'." And you could update your "scientific" understanding of the distribution of HP to account for the fact that precisely one character has 1000 HP.
The difference between my scenario and the video game one is merely quantitative: Pr(1000 HP | I'm not in a video game) < Pr(I'm a superlative | I'm not in a simulation), though both probabilities are very low.
Replies from: CstineSublime↑ comment by CstineSublime · 2025-01-07T03:54:10.960Z · LW(p) · GW(p)
I never said "falsified" in that reply - I said fake - a simulation is by definition fake (Edit: Yes I did, and now I see how I've been 'Rabbit Seasoned' - a simulation hypothesis falsifies this reality. I never said this reality is false. My mistake!). That is the meaning of the word in general sense. If i make a indistinguishible replica of the Mona Lisa and pass it off as real, I have made a fake. If some kind of demiurge makes a simulation and passes it off as 'reality' - it is a fake.
I've never heard of "anthropics" but I am familiar with the Anthropic Principle it's antecedents in pre-socratic philosophers like Heraclitus who are the first known record of the concept. Have you heard of Kant and German Idealism?
Indeed, your video game scenario is not even really qualitatively different from my own situation.
How? To take your example of being the tallest person: If all human beings were exactly 6 feet tall, and you were 600 feet tall, then you're saying that would be proof that you are in fact in a simulation. That might suggest you are in fact extremely special and unique, if you want to believe in a solipistiic, Truman-show style world.
if I were born with 1000 HP, you could still argue "data from within the 'simulation'...is not proof of something 'without'."
Yes. Exactly. I could. Although it would intuitively be less persuasive. But there aren't any 600 feet tall people in a world of otherwise uniform height.
The difference between my scenario and the video game one is merely quantitative: Pr(1000 HP | I'm not in a video game) < Pr(I'm a superlative | I'm not in a simulation), though both probabilities are very low
I don't understand where you're pulling that quantitative difference. Can you elaborate more?
Replies from: AynonymousPrsn123↑ comment by AynonymousPrsn123 · 2025-01-07T04:04:38.167Z · LW(p) · GW(p)
I don't appreciate your tone sir! Anyway, I've now realized that this is a variant on the standard Presumptuous Philosopher problem, which you can read about here if you are mathematically inclined: https://www.lesswrong.com/s/HFyami76kSs4vEHqy/p/LARmKTbpAkEYeG43u#1__Proportion_of_potential_observers__SIA
Replies from: CstineSublime↑ comment by CstineSublime · 2025-01-07T04:19:12.224Z · LW(p) · GW(p)
I didn't think there was anything off with my tone. But please don't consider my inquisitiveness and lack of understanding anything other than a genuine desire to fill the gaps in my reasoning.
Again, what is your understanding of Kant and German Idealism and why do you think that the dualism presented in Kantian metaphysics is insufficient to answer your question? What misgivings or where does it leave you unsatisfied and why?
I'm not immediately sure how the Presumptious Philosopher example applies here: That is saying that there's theory 1 which has x amount of observers, and theory 2 which has x times x amount of observers. However, "the world is a simulation" is but one theory, there are potentially infinite other theories, some as of yet unfathomed, and others still completely unfathomable (hence the project of metaphysics and the very paradox of Idealism).
Are you saying the presumptuous philosopher would say: "there's clearly many more theories that aren't simulation than just simulation, so we can assume it's not a simulation"
I don't think that holds, because that assumes a uniform probability distribution between all theories.
Are you prepared to make that assumption?
↑ comment by AynonymousPrsn123 · 2025-01-07T04:48:21.678Z · LW(p) · GW(p)
You are misinterpreting the PP example. Consider the following two theories:
T1 : I'm the only one that exists, everyone else is an NPC
T2 : Everything is as expected, I'm not simulated.
Suppose for simplicity that both theories are equally likely. (This assumption really doesn't matter.) If I define Presumptuous Philosopher=Distinct human like myself=1/(10,000) humans, then I get in most universes, I am indeed the only one, but regardless, most copies of myself are not simulated.
Replies from: CstineSublime↑ comment by CstineSublime · 2025-01-07T05:12:09.408Z · LW(p) · GW(p)
I'm still not sure how it is related.
The implicit fear is that you are in a world which is manufactured because you, the presumed observer are so unique, right? Because you're freakishly tall or whatever.
However, as per the anthropic principle, any universe that humans exist in, and any universe that observer exists in is a universe where it is possible for them to exist. Or to put it another way: the rules of that universe are such that the observer doesn't defy the rules of that universe. Right?
So freakishly tall or average height: by the anthropic principle you are a possibility within that universe. (but, you are not the sole possibility in that universe - other observers are possible, non-human intelligent lifeforms aren't impossible just because humans are)
Why should we entertain the possibility that you are not possible within this universe, and therefore that some sort of demiurge or AGI or whatever watchmaker-stand-in you want for this thought experiment has crafted a simulation just for the observer?
How do we get that to the probability argument?
↑ comment by AynonymousPrsn123 · 2025-01-07T05:18:43.749Z · LW(p) · GW(p)
I don't understand. We should entertain the possibility because it is clearly possible (since it's unfalsifiable), because I care about it, because it can dictate my actions, etc. And the probability argument follows after specifying a reference class, such as "being distinct" or "being a presumptuous philosopher."
Replies from: CstineSublime↑ comment by CstineSublime · 2025-01-07T05:36:21.054Z · LW(p) · GW(p)
We should entertain the possibility because it is clearly possible (since it's unfalsifiable), because I care about it, because it can dictate my actions, etc.
What makes you care about it? What makes it persuasive to you? What decisions would you make differently and what tangible results within this presumed simulation would you expect to see differently pursuant to proving this? (How do you expect your belief in the simulation to pay rent in anticipated experiences [LW · GW]?)
Also, the general consensus in rational or at least broadly in science is if something is unfalsifiable then it must not be entertained.
And the probability argument follows after specifying a reference class, such as "being distinct" or "being a presumptuous philosopher."
Say more? I don't see how they are the same reference class.