If you have been paying attention to pop culture and the Internet(s) lately, you’ll understand that there has been a flourishing new series of weird content .. There are various sources out there providing us some amazing artificial intelligence artwork. The type of art that enters your brain and is so vivid you wince, you look perplexed.. you get creeped out.
So let’s explain..
It is everywhere!! AI art! It is abundant.. it is booming fast. And so often it is dark, creepy.. obscure and weird..
People have embraced the weirdness..
John Oliver got fun and self-deprecating about AI art on his show..
We also got into the act several weeks ago and made our own AI art by asking a computer to create “Coal Speaker.” It was pretty odd, as evidenced by our favorite picture:
YOU HAVE OPTIONS..
Starryai.. Wombo.. Dall-E.. DreamStudio.. Midjourney.. You have options. AI is creating art for people on multiple platforms. As we have seen, the art normally gets dark and macabre quickly.. It is changing the world of art as well. Big time..
Think of it.. your imagination is contained to what it can conjure, what your brain can create.. but AI art can take it all in.. create something from nothing, or collect images and thoughts from various ‘Internets’ to bring a strange and bizarre sentence to life.
Search for “lima beans hunting aliens on the planet Poptart” .. and AI art will create it for you. Just for fun, we did that exact thing, and we came up with this (DreamStudio didn’t know quite what to do with it)
It all started with DALL·E … it made major waves since it was first revealed back in April 2022 because it was the most advanced AI art generator that we has been introduced..
Some artists are concerned. Other artists are just going hog wild creating images that are made in rapid fire by computers, based on random sentences.. random thoughts..
There is a huge new trend on TikTok of people search for each line of song lyrics to create an array of beautiful videos to famous songs..
All created in seconds..
All created with AI–no humans painting or drawing.. just artificial intelligence quickly changing our world…
BUT AI GETS CREEPY
So often, it appears that AI art goes down a rabbit hole of high strangeness.. For example, we asked for a series of images of “Pope Francis” and eventually AI art created what appears to be the Pope fighting off a double-demon version of himself..
Other websites have capitalized on AI art and created “the last selfie,” something that made its viral rounds in the summer. Most images of the AI created “last selfie” appeared to be Armageddon. The end times.. nuclear weapons.. This is dark stuff. AI is a downer!
WHAT IS THE OPPOSITE OF MARLO BRANDO?
Thought AI art was weird? Here is where things get … really strange.
Meet Loab. The woman, or creation from AI art, of your nightmare..
Let’s try to unpack ..
It all started back in September as we entered into the autumnal season.
An artist on Twitter under the handle @supercomposite used a technique called “negative prompt weights.” That means the user attempts to force an artificial intelligence system to generate the opposite of what they enter in the prompt.
Different terms can be “weighted” in a data set to determine how likely they are to appear in the results. You are simply telling AI to “Generate what you think is the opposite of this prompt.”
There is actually some coding that goes into this procedure..
So using this negative-weight prompt on the word Brando–with supercomposite referring to Marlon Brando–it created a logo image with a city panorama and the words “DIGITA PNTICS”.
At this point, supercomposite figured using using negative weights for the words DIGITA PNTICS it may see Marlon Brando..
No. Wrong. Instead, Loab was born.
The AI generator has not been named, but some have speculated that due to its quality it MidJourney or Stable Diffusion.
From the Tweet back in September:
This is what supercomposite went on to say about Loab, in a series of Tweets:
Since Loab was discovered using negative prompt weights, her gestalt is made from a collection of traits that are equally far away from something. But her combined traits are still a cohesive concept for the AI, and almost all descendent images contain a recognizable Loab.
The angel hallway + Loab also produced art with such copious gore that probably very few people want to see them, but here are two. I don’t feel comfortable posting the most disturbing ones, borderline snuff images of dismembered, screaming children.
There is something moving to me about these grotesque scenes and the desperation, panic, and sadness that they convey. Again, these are produced with other images as inputs, and no text. They are the result of “cross-breeding” images of Loab with images of other things.
The images that result from crossing Loab with other images can in turn be crossbred with other images. The AI can latch onto the idea of Loab so well that she can persist through generations of this type of crossbreeding, without using the original image.
Even when her red cheeks or other important features disappear, the “Loabness” of the images she has a hand in making is undeniable. She haunts the images, persists through generations, and overpowers other bits of the prompt because the AI so easily optimizes toward her face.
AND MUCH MORE..
So it can go from Loab:
To these other versions:
At this point, as she gets more famous, Loab could be potentially new “creepypasta” of the internet, but in the other side it could be a grotesque source of inspiration for making horror movies..
One of our favorite YouTube accounts, NEXPO, featured Loab in a recent video, and it is worth the viewing:
Matthew Skala, a computer scientist, tried to explain how Loab was created..
But why so dark.. Why does AI keep taking us to the abyss?
The AI generated images, derived from the original Loab image, go straight away into the realm of horror, graphic violence, and gore. But no matter how many variations have been made, the images all seem to depict the same horrifying woman.
They don’t change.. Loab stays.. She continues persisting through that weird and dark tunnel of AI art..
In the end, AI programs generate images using models that are trained on billions of images and are too complex to understand why a certain result is achieved.. So you will not really know why or how Loab, or any other oddity, was created. It just was..
Loab is haunting.
Will she replace Momo and Huggy Wuggy? She sure could..
It is the Halloween season as this post is written.. so we are getting creepy. HyperAllergic joins in the fun with showcasing some other dark and grim AI art..
….yes….AI art continues to be just be glum and dire.. distraught and dark.
Until you ask AI itself to create “Happy AI creating AI Art” … and we get this:
But before you get overjoyed by the colorful and happy image of AI showcasing itself, don’t forget what else it creates. Images of the vast unknown.. the sad. The tired. The horror.
Yes indeed… remember that before you embrace AI art..