AI-generated breakfast food
Used GPT-3 to generate new breakfast cereals and descriptions.
Some are a bit questionable.
I went on a quest to get CLIP to recreate those giant wooden playground complexes from my childhood. You know, one of these.
I figured it would work - they already look like a fever dream of an over-enthusiastic AI. I never got close.
Neural net generated food items to enjoy!
Pixray generating "View from the end of the Thanksgiving Table"
Demo here: https://pixray.gob.io/text2pixel/
Pixray Swirl uses CLIP to generate each frame to match my prompt, starting from a shifted version of the previous frame.
Come for the jellyfish, stay for the colossal jam jar.
After I gave it a few examples of changing one style of text into another, I got GPT-3 to draw on the vast knowledge encoded in its internet training and improve some common error messages.
If I give it a few examples, I can get GPT-3 to change the style of a sentence. See how it improves my writing!
I decided to see what CLIP+VQGAN would do when I gave it a house and told it to make it haunted.
vs modified to better match the phrase "a haunted Victorian house"
AI is based on math so it must be correct.
New AI Weirdness post! https://www.aiweirdness.com/clip-backpropagation/
It's easier to make an industrial demon than a unicorn cake? I experimented with @RiversHaveWings's new diffusion-guided CLIP method.
"industrial demon, matte painting, artstation HD"
"unicorn cake with golden horn and rainbow sprinkles"
There's no cake fail like an AI-generated cake fail.
I had CLIP+VQGAN attempt some classic showstopper cakes. Here's its "galaxy glaze cake"
Pretty striking the difference that a byline makes when asking CLIP+VQGAN to generate images.
"Internet infrastructure by James Gurney"
new post at aiweirdness.com!
Tried to generate a chair made of beetle kill pine.
It's a wood with distinctive blue streaks, salvaged from trees killed by pine bark beetles. Turns out it was easier to get CLIP+VQGAN to generate the texture than the chair.
Asking CLIP+VQGAN to generate
"a car driving down a road in monument valley"
versus "A car driving down a desert road in monument valley | dramatic atmospheric ultra high definition free desktop wallpaper"
SAME TRAINING DATA
What fascinates me about generating images with CLIP is that it CAN generate depth and realistic textures, but only if you know how to ask for them.
"A herd of sheep grazing on a lush green hillside" alone
vs with "amazing awesome and epic" added
The problem with taking startup advice from a text-generating neural network is even the advanced ones know more about what sounds good than about what actually works.
In which a neural net prompted with NOAA's Atlantic Basin Storm Name Pronunciations guide further erodes the already tenuous link between English spelling and pronunciation
Make your move while the AI-generated Bachelorette contestants are still available!
Generate buzz with these AI food truck concepts.
Remember, GPT-3 scanned most of the internet during training so it knows what's popular!
Botsplaining: turns out it's much easier to get a neural nets to sound confident than to actually be correct.
I train neural networks to write humor at aiweirdness.com | Also a research scientist in optics. She/her
The Wandering Shop is a Mastodon instance initially geared for the science fiction and fantasy community but open to anyone. We want our 'local' timeline to have the feel of a coffee shop at a good convention: tables full of friendly conversation on a wide variety of topics. We welcome everyone who wants to participate, so long as you're willing to abide by our code of conduct.