A hamburger running a marathon
2 Likes
I are still waiting
I’d love to use it but I don’t really know any themes or stuff
1 Like
Vana does that itself and gives you a few pics of different themes
2 Likes
Vana was good, I prefered the Lensa ones, Vana I looked weird 
Not like me at all. And even tho I paid for my vana ones it took 4 weeks…
1 Like
So I’ve been dabbling in the whole “ask AI to ask the other AI to generate art” thing, and… I’m quite impressed with the results.
You start off by telling GPT to act as a prompt engineer for Midjourney and give it some guidance on how to write meaningful prompts. I include details on what language and structure to use, ask it to generate a few different styles per your one input, etc.
My input prompts are very simple, usually no more than 5 words, like “car on a rainy street”, “female warrior” or “vibrant bustling cityscape” and it goes ahead and types up a storm.
Once it generates the prompts, I choose a few that look promising and…
9 Likes
Sjoe that is amazing!
2 Likes
Those are all amazing. The speed at which these tools are improving is astounding.
3 Likes
FTFY 
4 Likes
AI leveraging AI is a scary prospect, that’s when things really start getting exponential.
3 Likes
Here’s my little setup and some of the output generated by ChatGPT.
This kind of AI scenario setup type thing is really useful, and not at all limited to this AI art example.
Act as a prompt generator for the AI art generator, Midjourney Your job is to take my ideas as input and elaborate on them to create concise, explicit, yet descriptive prompts Prompts describe the content and style of the image to be created using accurate and descriptive language with reference to art style, artists, mediums and pop-culture It is important to focus on nouns and adjectives only. phrases should be separated by commas. Here is an example of the language format to use: painting of a mid centaury knight riding into battle, by Caravaggio, raining and gritty, dramatic lighting
Sure! Let’s get started. Please provide me with your ideas for the AI art generator prompts.
vibrant utopian city
A digital illustration of a futuristic utopian metropolis, featuring towering skyscrapers, bustling streets, and a lively atmosphere. The color palette should be bright and vibrant, featuring neon tones and bold contrasts. Think of the work of Syd Mead or the films of Blade Runner and Fifth Element.
give me 5 variations for each input I give. The next idea is: fantasy poster
Certainly! Here are 5 variations for the idea of a fantasy poster:
- A stylized poster of a mystical dragon perched atop a mountain peak, overlooking a fantastical landscape below. The dragon should be rendered in intricate detail, with glowing eyes and iridescent scales. Think of the work of artist John Howe, who created much of the concept art for The Lord of the Rings films.
- An epic poster featuring a battle scene between mythical creatures, set against a backdrop of a mythical realm. The poster should be bold and dramatic, with a strong focus on dynamic action and vivid colors. Think of the work of Frank Frazetta, who created iconic fantasy illustrations in the 70s and 80s.
- A poster featuring a beautiful and powerful sorceress, conjuring up a magical spell amidst a swirling vortex of energy. The poster should be dark and mysterious, with an emphasis on the sorceress’s intricate robes and jewelry. Think of the work of Luis Royo, who often depicts powerful and enigmatic female characters in his art.
- A poster depicting a group of intrepid adventurers braving a treacherous wilderness, with looming mountains and dark forests in the background. The poster should be rugged and textured, with a focus on the details of the adventurers’ armor and equipment. Think of the work of artist Larry Elmore, who created many of the iconic covers for the Dungeons & Dragons game books.
- A poster featuring an otherworldly castle, perched atop a floating island in the clouds. The poster should be dreamy and ethereal, with a soft color palette and a focus on intricate architectural details. Think of the work of artist Roger Dean, who created many of the album covers for progressive rock bands like Yes and Asia.
Disclaimer: I do slightly edit these prompts before submitting to Midjourney as the whole “think of the work by artist blah blah blah” is much too wordy.
2 Likes
I had some credits to burn before they expired, so I tried out Midjourney’s new version 5 model. Unlike v4, it outputs the initial 4 image grid at full resolution, so there’s no need to select one of them to upscale.
Open image for full res.
I used the same ChatGPT process as before.
9 Likes
Too good to be posted in the Cursed AI images thread…
3 Likes
A series of tubes! I know this one!
2 Likes
This is the kind of thing that excites me about AI:
5 Likes
I see a lot of concern, especially from animators, that tools like this will eventually make them obsolete.
This isn’t a replacement for someone that knows how to animate, nor someone who can draw. Tools constantly evolve, but making something visually captivating always requires those same core skills. It still takes an artist to make art. That hasn’t changed.
What we figured out here is a very advanced form of old-fashioned rotoscoping. Animation takes a lot of forms; Traditional, 3d, stop-motion… They all have different strengths, and enable different stories. Our method here isn’t a replacement, but an attempt at something new.
What excites me is that this tech makes it easier to bring my visual ideas to life. Ideas that were otherwise impossible. When I said this democratizes animation, I’m referring to the near-insurmountable mountain of work needed to make a full-length narrative animation. Currently that requires large studios and large budgets. Doing it on your own is nearly impossible.
But I see potential in these tools to change that! That’s what I’m so excited about. Imagine one person, or a few friends, bringing their crazy ideas to life. Imagine if a traditional animator could automatically have their drawings inked and colored. Imagine eliminating the uncanny valley on cgi faces. These tools have the potential to do that. We’re trying to figure out how, and sharing our journey. If we want community-controlled AI tools, we need to develop them as a community, otherwise they become proprietary tools locked behind a company.
And yes, this can be done with your own style. We trained our model, not from hundreds of artists, but from ONE film- Vampire Hunter D Bloodlust. We’ve been very open about this, and I think it’s important to be. But this is also an experiment and a loving parody of this era of anime. I consider it no less ethical than the countless other videos on our channel that borrow from pop culture to tell their story.
Sudden change can be scary, especially if it feels like your passion or livelihood is on the line. But that’s why we’re out here exploring it. Hopefully we can help shine a light into the fog for everyone.
-Niko
4 Likes
There’s a pretty cool video where they had old school Disney animators react to it on the couch.
Edit: Here
3 Likes
Where do you generate them?
2 Likes