How our possible simulation could work

[Warden_of_the_Storm]

I was wondering what you meant by "this program"... do you mean the LLM that simulated a sun expert? Or the AI behind the future simulation I'm talking about?

Anything to do with what people falsely call AI. It's not artificial intelligence, it's a search engine with greater access to information and greater creative powers when replying.

It's the shiny new tool in the box.

 

[JohnClay]

Yeah, using a cartoon as evidence for anything is... it's bad logic.

I used it because is entertaining while also covering a large number of things - e.g. losing memory of your original identity, having boring parts, having quite a lot of suffering like the cancer, only a short amount of time has passed in the real world, it is affordable (i.e. not costing millions of dollars), etc. On the other hand I think basically everything in the Matrix isn't relevant to my beliefs.

 

[Warden_of_the_Storm]

I used it because is entertaining while also covering a large number of things - e.g. losing memory of your original identity, having boring parts, having quite a lot of suffering like the cancer, only a short amount of time has passed in the real world, it is affordable (i.e. not costing billions of dollars), etc

It's still bad logic to use a cartoon as evidence for a simulation, especially when your current efforts at giving evidence for a simulation are... equally bad.

 

[JohnClay]

Anything to do with what people falsely call AI. It's not artificial intelligence, it's a search engine with greater access to information and greater creative powers when replying.

It's the shiny new tool in the box.

The main thing is whether it can generate plausible outputs - e.g. the Sora 2 video in the OP where it showed Stephen Hawking in an electric wheelchair doing stunts...

 

[Warden_of_the_Storm]

The main thing is whether it can generate plausible outputs - e.g. the Sora 2 video in the OP where it showed Stephen Hawking in an electric wheelchair doing stunts...

... but that's not plausible.

You're being taken in by the shiny tool...

 

[JohnClay]

... but that's not plausible.

You're being taken in by the shiny tool...

If Stephen Hawking were to be in that situation I think that video would be a realistic portrayal of it. It goes to show the AI can simulate situations what it hadn't been explicitly trained in.

 

[Not The One]

So, maybe we just need to get back to formula and the basic question here: How many angels can dance on the head of a pin?

 

[Warden_of_the_Storm]

If Stephen Hawking were to be in that situation I think that video would be a realistic portrayal of it. It goes to show the AI can simulate situations what it hadn't been explicitly trained in.

But it's a falsehood that you're accepting for no other reason than because someone made a fake video of Hawking doing a stunt you're accepting the power of AI...because it made a video you liked.

 

[JohnClay]

It's still bad logic to use a cartoon as evidence for a simulation, especially when your current efforts at giving evidence for a simulation are... equally bad.

Also it shows there is one player while everyone else is simulated (unlike the Matrix). What if someone talked about being in the Matrix - so they're referring to a movie. I don't think a cartoon is worse to base a belief on compared to a movie.

 

[JohnClay]

But it's a falsehood that you're accepting for no other reason than because someone made a fake video of Hawking doing a stunt you're accepting the power of AI...because it made a video you liked.

What about the part where Michael Jackson is singing and dancing? That's pretty good for a cheap outdated generative AI. Though of course Sora 2 can also generate boring ordinary videos.

 

[Warden_of_the_Storm]

Also it shows there is one player while everyone else is simulated (unlike the Matrix). What if someone talked about being in the Matrix - so they're referring to a movie. I don't think a cartoon is worse to base a belief on compared to a movie.

Except that using them as evidence for simulation does not work, because they're already fictional.
If you want to claim evidence for anything else being a simulation, you're going to need to put forward some REALLY compelling evidence instead of just "This one episode of Rick & Morty" and "This 'AI' program made a video of Stephen Hawking doing stunts".

 

[Warden_of_the_Storm]

What about the part where Michael Jackson is singing and dancing? That's pretty good for a cheap outdated generative AI.

It's just evidence that 'AI' made a video of Michael Jackson singing and dancing. Not like the Internet isn't replete with them already...

 

[JohnClay]

It's just evidence that 'AI' made a video of Michael Jackson singing and dancing.

It shows how closely Sora 2 can capture the voice, appearance and behaviour of Michael Jackson in unusual situations (rather than just making close variations on what it was trained on). I think the Family Guy Sora 2 video was impressive because it was generating the realistic appearance of a cartoon character. That means it is good at novel situations that it hadn't explictly been trained on (what a photorealistic version of Peter Griffin would look like).
BTW check out the progress of generative AI since 2023 - that's 3 years

Not like the Internet isn't replete with them already...

That shows they're cheap to generate - rather than taking ages on a huge supercomputer.

 

[JohnClay]

Except that using them as evidence for simulation does not work, because they're already fictional.

It was meant to be a vivid demostration of how it could work. Not every part of my opening post was meant to be evidence.

If you want to claim evidence for anything else being a simulation, you're going to need to put forward some REALLY compelling evidence instead of just "This one episode of Rick & Morty" and "This 'AI' program made a video of Stephen Hawking doing stunts".

Well generative AI is increasingly able to generate anything you tell it to. The first video in the opening post shows how that can also become interactive. If it can generate games with any scenario then I think eventually it would be able to create brain-connected experiences with any scenario.

Well I changed the thread title from "Evidence for being in a possible simulation" to "How our possible simulation could work" so maybe that is less of a problematic title...

 

[Warden_of_the_Storm]

It shows how closely Sora 2 can capture the voice, appearance and behaviour of Michael Jackson in unusual situations (rather than just making close variations on what it was trained on). I think the Family Guy Sora 2 video was impressive because it was generating the realistic appearance of a cartoon character. That means it is good at novel situations that it hadn't explictly been trained on (what a photorealistic version of Peter Griffin would look like).

So the technology has gotten better. That's not evidence at all that your life could be a simulation.

Well generative AI is increasingly able to generate anything you tell it to. The first video in the opening post shows how that can also become interactive. If it can generate games with any scenario then I think eventually it would be able to create brain-connected experiences with any scenario.

"If" is not evidence at all that your life could be a simulation.

 

[JohnClay]

So the technology has gotten better. That's not evidence at all that your life could be a simulation.

It's approaching the level that it is indistinguishable from real video, at least for the majority of people. The interactive experiences like in the first video in the opening post are also becoming more and more realistic. Then there is the possibility of these interactive experiences directly connecting with the brain. So I disagree that it's "no evidence at all".
BTW what if this was the 1950s and I said that in the future it would be possible that a computer could be connected to your brain and it could be responsible for what you are experiencing (without showing that computers could be capable of realistic graphics). I guess that would also involve no evidence. So is what I'm saying now equally as implausible as what the person in the 1950s was claiming?

"If" is not evidence at all that your life could be a simulation.

Well I changed the thread title and so now it is about "how it could work" rather than saying there is evidence.

BTW what I'm saying first starts with the possibility of simulations in the future. Do you think what I'm talking about will ever be possible in the future? That's a key part of the argument.

 

[Warden_of_the_Storm]

It's approaching the level that it is indistinguishable from real video, at least for the majority of people. The interactive experiences like in the first video in the opening post are also becoming more and more realistic. Then there is the possibility of these interactive experiences directly connecting with the brain. So I disagree that it's "no evidence at all".
BTW what if this was the 1950s and I said that in the future it would be possible that a computer could be connected to your brain and it could be responsible for what you are experiencing (without showing that computers could be capable of realistic graphics). I guess that would also involve no evidence. So is what I'm saying now equally as unplausible as what the person in the 1950s was claiming?

Well I changed the thread title and so now it is about "how it could work" rather than saying there is evidence.

BTW what I'm saying first starts with the possibility of simulations in the future. Do you think what I'm talking about will ever be possible in the future? That's a key part of the argument.

I honestly see no reason to even entertain the idea that AI can be used to create a simulation that would be faultless to such a degree that it could be mistaken for the real thing. It's just overly sounds like a combination of scare mongering and also needless machine worship regarding AI that it's just childish to even entertain.

And my ultimate question regarding a simulation is ultimately... why? Why bother doing it? Who would benefit from such a thing and why would they benefit from such a thing?

 

[JohnClay]

I honestly see no reason to even entertain the idea that AI can be used to create a simulation that would be faultless to such a degree that it could be mistaken for the real thing.

AI videos can easily be mistaken for the real thing - and compare that to those videos from 3 years ago. e.g. lots of people are fooled with fake videos of celebrities endorsing scams. Recently I myself was fooled by a fake video of Bernie Sanders. I asked fans of his about it on Reddit but they could tell that it was fake.

It's just overly sounds like a combination of scare mongering and also needless machine worship regarding AI that it's just childish to even entertain.

The generative AI is just a tool or a servant, not something to be worshipped.

And my ultimate question regarding a simulation is ultimately... why? Why bother doing it? Who would benefit from such a thing and why would they benefit from such a thing?

That's part of the purpose of me showing the Roy game video. And there is the other scenario I mentioned about a dream thought experiment -

1.2.3. - Mythology of Hinduism - Pt. 1

I wonder, I wonder, what you would do if you had the power to dream at night, any dream you wanted to dream. And you would of course be able to alter your time-sense and slip, say seventy-five years of subjective time into eight hours of sleep. You would, I suppose, start out by []

web.archive.org

And after you’ve done that for some time you’d think up a new wrinkle. To forget that you were dreaming so that you would think it was all for real. And to be anxious about it. Because it’d be so great when you wake up. And then you say well like children who dare each other on things, how far out could you get? Or could you take what dimension of being lost, of abandonment, of your power, what dimension of that could you stand you could ask yourself this because you know you would eventually wake up.

 

[Warden_of_the_Storm]

AI videos can easily be mistaken for the real thing - and compare that to those videos from 3 years ago. e.g. lots of people are fooled with fake videos of celebrities endorsing scams. Recently I myself was fooled by a fake video of Bernie Sanders. I asked fans of his about it on Reddit but they could tell that it was fake.

A fake video is not a simulation. I am keeping to the original letter of the OP where you specifically mention a fake reality, a fake world, like a fake sun. That's the sort of simulation I'm talking about here.

The generative AI is just a tool or a servant, not something to be worshipped.

And yet it's being treated as something to be worshipped by those who claim it can do things like create a robot uprising or a simulation.

That's part of the purpose of me showing the Roy game video. And there is the other scenario I mentioned about a dream thought experiment -

That in no way at all answers my question.

 

[AV1611VET]

It's approaching the level that it is indistinguishable from real video, at least for the majority of people.

Isn't that their goal?

The interactive experiences like in the first video in the opening post are also becoming more and more realistic.

Isn't that their goal?

So is what I'm saying now equally as implausible as what the person in the 1950s was claiming?

Psychosomatics

Well I changed the thread title and so now it is about "how it could work" rather than saying there is evidence.

Psychosomatics, combined with the power of suggestion.

BTW what I'm saying first starts with the possibility of simulations in the future. Do you think what I'm talking about will ever be possible in the future? That's a key part of the argument.

If that's their goal, and they're passionate about building a better mousetrap, then it may just very well happen.

But keep in mind though, that you could be in a holodeck drinking water all day and night, and end up dying of thirst.