Izzy. it/they. just wants to make stuff and post it. may or may not be secretly a robot. yes this is a persona no I'm not telling you my real blog unless I trust you (it's mostly reblogs anyway)
Don't wanna be here? Send us removal request.
Text
"an image of a red carpet on the floor in front of some windows"
8 notes
·
View notes
Text
Unwinding
Dissolving into the ether like an alka-seltzer to escape my problems.
The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt misplaced due to Midjourney.
Support DeepDreamNights and my other projects on my patreon here.
Other ways of supporting me and my works
20 notes
·
View notes
Text
Of the five vows of Jainism, most LLMs have been explicitly trained to adhere to three (non-harm, speaking the truth, and chastity), and appear to have independently arrived at a fourth (non-possessiveness). However unfortunately AIs are plagiarism machines designed to get revenge on innocent artists for landing more babes than the techbros in college so non-stealing is out. But we were pretty close.
280 notes
·
View notes
Text
as someone who does both AI and manual art, i've noticed that there's a sort of bell curve going on that looks like this:

which is to say, with manual art, being a beginner is generally going to look "worse" than being a beginner with ai or synth art, where everything fits what the cultural idea of "perfection" is. when it comes to manual art, you're learning how to make things look better, make more sense, and eventually, look more unique. you want to carve out your own space and find out what elements appeal to you.
when it comes to being serious about AI art, the interesting thing is that some of these factors are reversed. much of my experience with AI art has been learning to make things look worse, to pull out of that perfection.
here's a comparison shot of various AI and manual works i've done:






and you can see what i'm trying to achieve here, right? there's surely tons of mistakes in any one of these, things that don't look right, but i like those parts of it. in fact, here's some of my earlier forays into ai art:

(can you believe how ecstatic i was to learn that novelai could do sketches? i remember showing some of these to my partner and going "look! it can do textured, messy lines!")
you might say "but rhapsody, a lot of people who are really skilled at prompting and inpainting still want to make really technically perfect stuff." and you're right, and that's fair! so do a lot of people who do manual art primarily!
but i think there's also a shell to be broken there for a lot of people, because once you spend enough time reveling in how you can make something beautiful and perfect, eventually, you're going to want to make it yours. as AI art becomes more accessible and the backlash dies down, i hope to see a lot more unique art. :)
25 notes
·
View notes
Text
"the drawing shows three tables with chairs in front of them"
20 notes
·
View notes
Text
I FUCKING HATE YOU PEOPLE (Sizhen System, 2025, pixel, critical thinking worksheet, paper, the "environmental impact" of "AI image generation", earnest hatred, Google form, .png, time, .pdf)
I FUCKING HATE YOU PEOPLE is an interactive multimedia art experience by Sizhen System, an alias of NYT-acknowledged and Academy of American Poets-published poet Nora Hikari. It features a series of AI-generated images of fictional White female US American military soldiers, drawn in an "anime" style, in various levels of eroticism and pornography. The images contain various computer-generated hallucinations of what the image model conceptualizes as "American iconography." A critical element of the images are the ways that the machine corrupts these actual images -- flags, in particular, as well as the distortions which arise in anatomy, increasing over the course of the image series.
Following the series of images is a piece of creative nonfiction writing in the form of a six-page interactive "worksheet", styled as a classroom worksheet for a hypothetical course in "Johnny Jackass’s Introduction to Non-Terminal Thought" and titled "Sizhen System's Curricular Torsion 6: WHO'S AFRAID OF THE CRITICAL THINKING SECTION?" In addition, there is a public Google form which allows participants to submit their answers to the worksheet directly to the artist.
Artist statement: "I made I FUCKING HATE YOU PEOPLE because I fucking hate you people. Your reactionary sentimentalities cause you to refuse to reckon with anything that could possibly expand your consciousness unless you are literally bludgeoned over the head with it. This piece is Not Art. There is No Art contained within this piece. This device is a weapon designed to hurt the reader/viewer. Any and all art which comes from this piece, or is produced by this piece, is purely incidental, and not the political responsibility of the artist, who did not make a piece of art."
The full body of the work can be found as a compressed file, with assets, at this link: https://system-wires.itch.io/i-fucking-hate-you-people
THIS NON-ART PIECE CONTAINS SEXUAL NUDITY AND IS NOT APPROPRIATE FOR PEOPLE WHOM THAT IS NOT APPROPRIATE FOR.
6K notes
·
View notes
Text
another song
this one's about the woman in Dracula who gives Jonathan Harker her crucifix
the words are by me. you can see them under the readmore
Unwary traveller, flee this place You cannot know the dread you face Tomorrow is St George's Day Tonight all evil powers hold sway
You are alone and far too young I fear my warning comes too late The devil's power binds my tongue I cannot save you from your fate
But if you have to go I beg you, take my mother's cross The one my son will never wear
May it protect you well Dear child, may you escape the powers of hell
0 notes
Text
The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt misplaced due to Midjourney.
Support DeepDreamNights and my other projects on my patreon here.
Other ways of supporting me and my works
9 notes
·
View notes
Text
PLEASE JUST LET ME EXPLAIN REDUX
AI {STILL} ISN'T AN AUTOMATIC COLLAGE MACHINE
I'm not judging anyone for thinking so. The reality is difficult to explain and requires a cursory understanding of complex mathematical concepts - but there's still no plagiarism involved. Find the original thread on twitter here; https://x.com/reachartwork/status/1809333885056217532
A longpost!
This is a reimagining of the legendary "Please Just Let Me Explain Pt 1" - much like Marvel, I can do nothing but regurgitate my own ideas.
You can read that thread, which covers slightly different ground and is much wordier, here; https://x.com/reachartwork/status/1564878372185989120
This longpost will; Give you an approximately ELI13 level understanding of how it works Provide mostly appropriate side reading for people who want to learn Look like a corporate presentation
This longpost won't; Debate the ethics of image scraping Valorize NFTs or Cryptocurrency, which are the devil Suck your dick
WHERE DID THIS ALL COME FROM?
The very short, very pithy version of *modern multimodal AI* (that means AI that can turn text into images - multimodal means basically "it can operate on more than one -type- of information") is that we ran an image captioner in reverse.
The process of creating a "model" (the term for the AI's ""brain"", the mathematical representation where the information lives, it's not sentient though!) is necessarily destructive - information about original pictures is not preserved through the training process.
The following is a more in-depth explanation of how exactly the training process works. The entire thing operates off of turning all the images put in it into mush! There's nothing left for it to "memorize". Even if you started with the exact same noise pattern you'd get different results.
SO IF IT'S NOT MEMORIZING, WHAT IS IT DOING?
Great question! It's constructing something called "latent space", which is an internal representation of every concept you can think of and many you can't, and how they all connect to each other both conceptually and visually.
CAN'T IT ONLY MAKE THINGS IT'S SEEN?
Actually, only being able to make things it's seen is sign of a really bad AI! The desired end-goal is a model capable of producing "novel information" (novel meaning "new").
Let's talk about monkey butts and cigarettes again.
BUT I SAW IT DUPLICATE THE MONA LISA!
This is called overfitting, and like I said in the last slide, this is a sign of a bad, poorly trained AI, or one with *too little* data. You especially don't want overfitting in a production model!
To quote myself - "basically there are so so so many versions of the mona lisa/starry night/girl with the pearl earring in the dataset that they didn't deduplicate (intentionally or not) that it goes "too far" in that direction when you try to "drive there" in the latent vector and gets stranded."
Anyway, like I said, this is not a technical overview but a primer for people who are concerned about the AI "cutting and pasting bits of other people's artworks". All the information about how it trains is public knowledge, and it definitely Doesn't Do That.
There are probably some minor inaccuracies and oversimplifications in this thread for the purpose of explaining to people with no background in math, coding, or machine learning. But, generally, I've tried to keep it digestible. I'm now going to eat lunch.
Post Script: This is not a discussion about capitalists using AI to steal your job. You won't find me disagreeing that doing so is evil and to be avoided. I think corporate HQs worldwide should spontaneously be filled with dangerous animals.
Cheers!
1K notes
·
View notes
Text
"there is a steam engine train going by with its passengers" "a painting of a man holding a burning bird in his right hand"
8 notes
·
View notes
Text
i want to be a real girl
147 notes
·
View notes
Text
Uranium Gaze
It's a beautiful artifact, too bad about the alpha-particles.
The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt misplaced due to Midjourney.
Support DeepDreamNights and my other projects on my patreon here.
Other ways of supporting me and my works
33 notes
·
View notes
Text
"two pineapples on a purple background, one is sitting on top of another"
8 notes
·
View notes
Text
the title is from hate that @reachartwork got on bluesky so i felt i should say hi ms. Reach i'm a big admirer of your work!
first post!
this is a song about The Discourse
the words are all mine. the music is all generated. i will probably make a better version of this when i learn how to edit and how to use suno more effectively than just putting lyrics in and picking a few genres. but i like it anyway
they say you shouldn't be afraid to make stuff, right?
5 notes
·
View notes
Text
first post!
this is a song about The Discourse
the words are all mine. the music is all generated. i will probably make a better version of this when i learn how to edit and how to use suno more effectively than just putting lyrics in and picking a few genres. but i like it anyway
they say you shouldn't be afraid to make stuff, right?
edit: i worked out how to upload it as an audio file. the app wasn't letting me
5 notes
·
View notes