Give a prompt and an AI model can spit out breathtaking art, or something that completely misses the mark. What is this AI art world and how can you try it?
Artists around the world aren’t exactly at risk, but if you take a look at the world of computer-created art from AI models, you can see why there might be cause for concern. worry.
If you’ve been scrolling through Twitter, Facebook, Instagram, or any other social media platform in the past few months, chances are you’ve seen some unusual AI-created artwork popping up. It’s one thing, and usually happens from a small section of text that is run through by a computer program, which in turn provides a result in an image.
Images created by models and artificial intelligence applications are becoming a thing of today, and although they are not yet in everyone’s hands, anyone can start dabbling, providing a little text and make an image appear, almost by magic.
But it’s not quite magic, even if it is, as art generators and image makers using artificial intelligence use the technology to create images almost seemingly from scratch.
What is an AI Art Generator
When an artist makes a work, it is with an idea and tools. Whether these tools are physical media – paint, pen, charcoal, camera, etc. – or digital equivalents, a human creating a work of art typically takes hours, days, weeks, or even longer. They come from an idea and a talent, and are designed for a purpose, whether it’s an artist or someone just dabbling.
Computerized art generators are quite different. They are built through code and analysis, comparing many different images to text and building understanding. It’s a complex assortment of patterns and processes, ranging from text to images, using machine learning to understand prompt elements and join the dots.
In short, these systems in many ways try to mimic the workings of the brain by creating a neural network to understand text and make images out of it.
Even if you don’t understand what’s going on – and it’s so complex that it would probably apply to most people – the results are often amazing, because each dot of text tends to create something so whimsical and fantastic that it’s hard to believe a computer created it in the first place.
Text-to-image AI art makers are Disneyland for your imagination, where anything you want becomes an intriguing outcome, many of which bring the idea to fruition, though some sometimes don’t often have hilarious consequences.
Examples you might want to see include multiple artists running a theme, like what we did with Disneyland.
Above you can find examples (from left to right) of what Disneyland would look like when painted by Arthur Streeton, Wassily Kandinsky, Vincent Van Gogh, and even a version of Disneyland as if the park image was painted for death metal album covers. . Fun.
The results are often weird and fun, and usually very playful, but they can also be beautiful and make you want to print them out, throw them on your wall as if you’ve commissioned a beautiful piece of art for your living room or office.
It’s not so much the result of art, but of machine learning where technology learns what goes together and why, providing a wealth of answers and letting you shape or decide the outcome.
Which AI art generators can I play (and how)?
Although the technology seems remote and advanced, it is not beyond everyone’s reach. Depending on your comfort level with a web browser or chat app, you can now jump into AI Art Creators, giving you something fun to do when you have time to spare, or even at any time.
The most advanced art builder of the bunch, OpenAI’s DALL-E 2, is largely out of bounds for most, even as the beta builds and opens gradually. Google has a variant on the way in Imagen, but it might not even be used by the public.
However, there are options you can try just by using an app or visiting a website. Many of them are free and you can see instant results.
For each of these examples, we’ll run through the “Disneyland by Kandinsky” prompt to see what’s going on, so you can see the differences between them.
One of the more obvious choices to take doesn’t require a Discord account and can simply be used by visiting a website.
Craiyon, formerly DALL-E Mini, is a text-to-art generator that lets you insert a theme and see the results, usually in a grid of 6-9 variations.
These images are usually low res in Craiyon, and it won’t give you the cool high res images like some other AI art makers can, but at least you can give it a try and get a taste of what’s going on .
However, Craiyon is totally free, so you can just visit, try and see the results.
The one we had the most fun with is Midjourney, and it’s the one you’ll need the Discord app to play with, head to the Midjourney website, and join the beta. Once joined, all you have to do is jump into the Midjourney Discord server and find an empty channel for newbies to enter your prompt.
They can be as simple as “Miles Davis looking at his trumpet” to “Miles Davis looking at his trumpet under a solitary lamp on a rainy night, Voigtlander Super Nokton 29mm f/0.8, Cinelux ASA 100, medium format print”, identifying a complex subject , the camera and lens with which it was captured, the film and the resulting impression.
All of these variables could create a painting or even an image that looks like a photo, as Midjourney uses its newly updated algorithms to create fun, exciting, and genuinely compelling art.
To date, he has produced our favorite example of the test prompt for this article, “Disneyland by Kandinsky”. He produced most of the examples we use throughout this article.
However, Midjourney costs money, providing you with sample images to start with, around 15, but then charging you a monthly fee starting at $10 per month for around 200 images per month.
Named after a work by Van Gogh, Night Cafe is like Craiyon in that it lets you type in a prompt and see the artwork come out of it.
Night Cafe offers two types of AI art generation, text to image being the most popular, but there is also “style transfer”, allowing you to move styles from one image to another, as if you wanted Van Gogh’s swirling stars to be transferred to something else.
The text prompt is probably the one you’ll be playing around with, and it works by typing in a prompt and choosing a style, as well as deciding on an algorithm to use.
It should be noted that Night Cafe is free for the first five images, but after that you will need to spend one credit per image, either paying monthly from $10 per month or buying a pack of credits from $8 USD for 40 credits that you can use each time.
Dream of Wombo
An AI art maker available both on the web and through an app, “Dream” by Wombo features a text prompt similar to the art style, with a choice of art style to use, and even images to draw inspiration from. .
Interestingly, you can also turn these AI art creations into NFTs if you want. Whether they’re actually worth anything is a question for the frustratingly complex world of crypto value and image – our guess probably isn’t – but it’s an option we’ve never seen. on any other AI art generation platform yet.
The Wombo results are pretty cool too, and they’re also pretty free (at press time), though you may be able to order a print of your Wombo art, and that will cost some money.
How to have fun with an AI art generator
No matter what you choose to try, you can have fun with any AI art maker just by providing a prompt and then watching the magic happen.
And “magic” is probably the best word for it, because just like Arthur C Clarke said “any sufficiently advanced technology is indistinguishable from magic” and people have interpreted it, good luck figuring out exactly what exactly the IA for your invites to turn it into art. It’s technology, sure, but it might also be magic.
Is there a risk that AI art creators will erode what artists do?
While there’s always a risk that AI art will make the art of human artists less important, we’d say it’s more for fun than creating physical art, and for at least two reasons.
1. The art of a true artist is usually more than flat
When you get an artwork, it may not be an A4 or A3 flatbed digital print. It will likely be larger and possibly textured, with real brush strokes and a sense of physics that digital prints might not have.
2. You Can’t Make AI Art Prompts Exactly What You Want
The other point is that while cool, fun, and playful, AI art creators never do exactly what you want them to, at least not in these early stages.
You can describe whatever you want in more detail than just a prompt, adding any additional details you want there, sizes, technology uses such as “raytracing”, “HDR”, and “8K”, and even set the aspect ratio, but you’re still relying on a computer receiving instructions and a model trying to figure out what you want. It’s different from being able to talk to an artist and hear their story, how it was concocted, etc.
Examples above of Midjourney with themes created by Ms. Five (from left to right): Princess Penelope Elsa Dongo the Beagle, a rainbow woodpecker, a goldfish with wings eating cake, fairies dancing under a pink moon and Elsa as a unicorn princess.
There are other reasons why digital art may not affect the world an artist lives and works in, but these two reasons strike this writer and artist (with a fine art degree!) as two of the more obvious.
It should be noted that while the art of AI is not the art of a human, they are both art. Whether or not they deliver what you expect from your art will depend on who created it and who is viewing it.
And if you derive pleasure from it by creating something that would never exist, that’s fine. But if you’re looking for a great image to decorate your wall with texture and story, chances are you want to hire a real artist to do it for you, or buy something they already have.