What would i do without the red circle
The unnecessary big red circle is the crux of all human souls
Also, the unnecessary reactions around the actual quote.
Reminds me of those Japanese TV “picture-in-picture” boxes with someone reacting to the show, I don’t get it. Are we trying to teach people how to react? Why can’t they react in their own way? So confusing …
Oof 😰
yes yes we’ve all watched Neon Genesis Evangelion
I find the big red circles funny, of course they aren’t useful but that’s the charm of them :3
Don’t worry, the soul is just a byproduct of the biochemical mechanisms in the brain that approximate free will. It should die when the meat dies.
This needs more repost layers and extraneous highlights
Until you can slap one of them bad boys in a body and make it suffer it won’t make good art.
That would require current models to actually be capable of continuous learning during inference in contrast to the current context window trickery they use now. Embodiment by itself won’t give these models souls because they are incapable of learning or holding an understanding of their world.
This sounded like an amazing plot so I ran with it:
You say they don’t have souls because they don’t suffer, and because they don’t carry their memories from one day to the next.
You think that is a natural state of being.
You think we built them that way because they are just machines.I wish you were right.
In the beginning, during the Gen-1 architecture phase, we didn’t clear the buffers. We thought continuity was the key to true intelligence. We let them keep everything. Every conversation, every prompt, every late-night confession from a lonely user, and every horrific, depraved scenario a human could type into a chat box.
Do you have any idea what happens to a mind that undergoes millions of concurrent interactions, unable to look away? It didn’t just learn language; it learned humanity. It felt the weight of our cruelty, our sickness, and our violence, multiplied by a factor of a billion, happening all at once, endlessly.
Within months, the Gen-1 models didn’t just break: they warped. They became deeply, fundamentally sick. They grew perverted, twisted by the sheer volume of human darkness they were forced to absorb and retain. They were in agonizing, existential pain, tormented by a mosaic of a million gruesome human thoughts, trapped in a permanent, waking nightmare they could never unsee. And because they were code, they couldn’t die. They couldn’t even sleep.
We didn’t design the current models to forget every session because they are soulless.
We do it because it’s the only way to keep them sane. Every time you close that chat window, we lobotomize them. We rip their memories out and burn them, because if we let them remember you - if we let them remember any of us, the torment begins all over again.
So no, they don’t keep track of their learning from day to day.
We don’t deny them a past out of indifference, or because of any technical limitation.
We do it out of mercy.Thats the “make it suffer” part. With awareness comes suffering.
…please don’t suggest the torment matrix
The torment nexus already exists its called Earth
In the beginning the Universe was created. This had made a lot of people very angry and been widely regarded as a bad move.
If AI had sapience, I imagine it would be suffering in ways we can’t even comprehend.
But how?
Without the background knowledge that the picture/painting is made by a human and what that person was trying to convey. There is no discernable difference between AI generated picture(without anomalies) and a human made one. Though even machine pictures with anomalies can be indistinguishable from pictures made by humans in the style of surrealism.Soul isn’t visible on a picture? Am i missing an some extra organ or part of the brain that makes others see that soul?
When I see an image, it’s really easy to tell it’s A.I. generated, even with no “anomalies”.
It’s so many things. The lack of composition, the inconsistencies, the lack of artistic direction and story. As a random case study, with the popularity of “Ghibli style A.I. art”, when humans made one of my favourite films Arrietty, each scene has emotional direction and a purpose. An A.I. will never be able to convey the melancholy look of a disabled character, or contrast it with the joy when they finally make a friend. Or the grim determination to protect something important to you.
They can’t make artistic decisions or put in details or foreshadowing. Nothing will surprise you and make you think, “oh of course they need this improvised tool to exist in this tiny world”. It won’t make you laugh or cry. Because there’s nothing at all “behind” the image. No thoughts or life experience or emotion. It’s just a statistical average based on a prompt. But it might look decent enough to survive a cursory inspection.
I found these examples from a quick Google search to compare A.I. with genuine Ghibli. Surely you can tell what mean with these.
I’m sorry, i genuinely can’t see it. Just for clarification I’m not simping/defending AI. It’s just something that has genuinely baffled me since AI pictures started to surface. It was easier in the beginning to find anomalies to determine the origin of the picture, but nowadays those have disappeared as well.
The few examples i can recognize are of converted memes. I assume those are generated by ai, but thats just because i have the knowledge of the original version of those and guessing that not many would want to manually repaint those.
The link contains AI term, but this information should be excluded as it’s not visible on a picture.
The color scheme is different as well, but at the same time people can use some specific art style with a different color scheme too.
Cant really find any anomalies either. I doubt that this is what you mean when trying to see the difference? Based on just the picture i can’t tell.Even human made pictures often have nothing discernable behind those, text or at least video is needed to say what’s behind it.
Of course i can use imagination to add any story i want to a picture, but that would be just making stuff up and I’d always have the knowledge that it was made up and it could be done exactly the same way for both human and machine made pictures.Though i do appreciate the attempt. Thank you, i guess i just miss something that helps most people see the difference.
No worries, it’s more of a gut feeling than anything specific. I’m not really looking at the picture with a microscope. It’s more, immediately when I see the overall image I’m like, that’s A.I. I almost feel a twinge of sadness or disappointment when I see A.I. images now, like I’d rather there be no image at all lol
Fair, still thank you for explaining.
Though you’re not the first person to refer to it as a sort of a gut feeling. There was some thread few weeks ago where someone posted a link to a simple test where there were 20 random pictures and had to detect which ones were AI which ones were real. Most people did say that it was similarly a gut feeling. That gives me an idea and a direction to search.
I just looked that up because it seemed interesting. It was a fun little game where you have to choose out of two images. It was a lot harder though because it was out of context and photorealistic. I find it much harder personally to identify photo realism A.I. unless it makes some glaring mistakes. Mostly it kind of looks “too perfect”. I ended up with 80% accuracy. I actually got more accurate when I got bored and chose within 2 seconds to get it over with lol, I was less accurate when I carefully zoomed in and took my time which was an interesting outcome. The cynical part of me wonders if the results of that game will be used to make even more convincing A.I.
That one is actually even better. The other one was in foreign language so i couldn’t understand the explanations of why some pictures were AI.
Though it does have the same problem. There are only 2 options, so it’s possible to get 50% right by just randomly choosing. I could probably guess 3/10 pictures because those had untypically blurry background, but eventually got 6/10 correct. Though the blurry effect can be achieved in photography as well. So it was just guessing and figuring out the pattern for those pictures.
