NVIDIA's AI Image Generator and the Platonic Realm

NVIDIA's new AI Image generator is VERY impressive. Check it out:



The fact that AI is this good has a lot of interesting and possibly scary implications.

But aside from that this provides a good analogy for the "Platonic Realm of Forms". The Platonic realm is a semantic realm... a network of weighted connections just like the neural network that generated these images.

I've imagined before the "platonic coffee mug" and thought of a slider that could phase through all the different instantiations of a "coffee mug"... not unlike the sliders on this AI program that can phase through different instantiations of a human face or cat or a car or a bedroom.

Maybe some glitches in the matrix and Psi related phenomena are somebody bumping the slider on the reality level above this one.

Another thing to think about from the sociological perspective... as the next generation grows up with the knowledge that everything they see could be fake and that we could be living in a simulation... what does that do to public trust in authorities and institutions? Will the next generation become radical skeptics, solipsists, and rebels? Or will the fear of uncertainty and the inability to find solid ground to stand on make people more vulnerable to cults and authorities that promise security and solidity and truth?

Hmmm I don't think neural network weights are really semantic -> They're probabilistic weights adjusted by volume of data. They have no meaning in themselves.

That said I'd agree that if AI can produce so many accurate simulated figures it does lend consideration to the idea we are in a Simulation. Not to say it's evidence, but it does make one realize how a more advanced civilization could produce highly complex systems.

More than Psi, I'd go with Arvan's P2P simulation that QM and it's collection of oddities seems to align well w/ what we'd expect from a Peer to Peer simulation.

I've imagined before the "platonic coffee mug" and thought of a slider that could phase through all the different instantiations of a "coffee mug"... not unlike the sliders on this AI program that can phase through different instantiations of a human face or cat or a car or a bedroom.

I'm a Platonist or at least lean that way, but for the sake of humor interestingly enough it's said that cup was the focal point of Diogenes critique of Platonism:

Plato was discoursing on his theory of ideas and, pointing to the cups on the table before him, said while there are many cups in the world, there is only one `idea' of a cup, and this cupness precedes the existence of all particular cups.


"I can see the cup on the table," interupted Diogenes, "but I can't see the `cupness'".

"That's because you have the eyes to see the cup," said Plato, "but", tapping his head with his forefinger, "you don't have the intellect with which to comprehend `cupness'."

Diogenes walked up to the table, examined a cup and, looking inside, asked, "Is it empty?"

Plato nodded.

"Where is the `emptiness' which proceeds this empty cup?" asked Diogenes.

Plato allowed himself a few moments to collect his thoughts, but Diogenes reached over and, tapping Plato's head with his finger, said "I think you will find here is the `emptiness'."

Hmmm I don't think neural network weights are really semantic -> They're probabilistic weights adjusted by volume of data. They have no meaning in themselves.

That said I'd agree that if AI can produce so many accurate simulated figures it does lend consideration to the idea we are in a Simulation. Not to say it's evidence, but it does make one realize how a more advanced civilization could produce highly complex systems.

More than Psi, I'd go with Arvan's P2P simulation that QM and it's collection of oddities seems to align well w/ what we'd expect from a Peer to Peer simulation.


I'm a Platonist or at least lean that way, but for the sake of humor interestingly enough it's said that cup was the focal point of Diogenes critique of Platonism:

Plato was discoursing on his theory of ideas and, pointing to the cups on the table before him, said while there are many cups in the world, there is only one `idea' of a cup, and this cupness precedes the existence of all particular cups.


"I can see the cup on the table," interupted Diogenes, "but I can't see the `cupness'".

"That's because you have the eyes to see the cup," said Plato, "but", tapping his head with his forefinger, "you don't have the intellect with which to comprehend `cupness'."

Diogenes walked up to the table, examined a cup and, looking inside, asked, "Is it empty?"

Plato nodded.

"Where is the `emptiness' which proceeds this empty cup?" asked Diogenes.

Plato allowed himself a few moments to collect his thoughts, but Diogenes reached over and, tapping Plato's head with his finger, said "I think you will find here is the `emptiness'."

I’ve been using a “cup” for years to demonstrate the idea of platonic forms and had no idea Plato used a cup as well...

NVIDIA has another video out showing their incredibly credible driving simulator.

As you said... neural nets are probabilistic... and isn’t that the reality we find ourselves in?

I think that a neural net is not merely analogous to the platonic nature of reality but is how reality is created.

Neural nets were created to perceive and define but in defining, one creates, and now they are being used to dream up incredibly complex novel realities.

The more I learn about neural networks the more marks of neural nets I see in the world.

So there is a god and he has a brain!

As you said... neural nets are probabilistic... and isn’t that the reality we find ourselves in?

Well we have a reality where we cannot always predict what happens next, and when we do use prediction we are using induction.

But is that probabilistic in the way neural nets are? This seems different to me because reality has mysteries of causation but neural nets merely modeling reality without needing to provide any motive force.

Also neural nets - AFAICTell - are using probability as an expression of ignorance, similar to the probability of 4 out of 100 electrons bouncing of a reflective layer says nothing about why the numbers work out that way?