The year is 2022. In a post-pandemic world, robots are taking over. Okay, that might be a little bit drastic, but as we move forward into a future with smarter and more intelligent AI, robots are more intelligent than ever. In fact, robots are reaching the level where they could take over some of our jobs and automate a lot of what is traditionally ‘human’ work.
We all knew computers were (one day) going to help ease the load and take a lot of jobs off our hands. And now we have robot-filled factories building our vehicles, automated checkouts in almost every supermarket and even robot waiters in some restaurants! But one of the last places you might expect to see AI thrive? That would be the art sector…
‘Quirky little Japanese shops, buildings, store fronts, market street, dense, high detail, mid-day, overcast lighting’ via Midjourney.
Computer generator artwork explained
What we’re talking about here is AI image generators. These are computer programs where you can describe any image (or artwork idea that you’ve produced), and the computer then creates an image based on your words. And this is done in a matter of seconds!
You can be as brief or as descriptive as you like, and the results can be spectacular, or equally horrifying, depending on your standpoint on the matter…
What does this art entail?
The art created by these programs can be truly mind-blowing; it could easily rival work created with a more traditional human touch. Not to say that this is an easy feat to master; the trick in creating stunning visuals on this software lies in the users’ choice of words.
This includes using specific terms relating to the style of the work you’d like to create. Details on the environment and textural information, all the way to artistic inspirations can influence the end result greatly. And using the correct language here is a skill in its own right!
Varying styles of work can be created from a few simple words, including paintings and cartoons as well as photo-realistic images. The only limitations are what you can imagine…
‘Giant Mayan pyramid in the jungle, beautiful moody light, storm with lightning’ via midjourney
So, what does this mean for artists and illustrators alike? Could this be the end, should we all be brushing up our skills in a new field, and re-working our CVs?
You’ll be pleased to hear the answer is no, probably not (for now). While the software is capable of creating masterpieces, it can be hard to align the artists’ vision with how the computer interprets our words. Because the outcomes can be very random, and often totally different to what you’d hoped for.
AI artists aren’t fully capable… yet
For every beautiful image created, there are probably hundreds that aren’t quite right. The software can suffer from slight noise issues, strange, warped areas and ‘uncanny valley’ syndrome when creating realistic faces. Still, we can use these programs to create more simplistic requests accurately or even to help build inspiration and bring ideas together.
‘Man full body with fire in space, NASA’ via midjourney
As displayed in some of the above examples, pieces of work created by AI can be brilliant. The process? It just involves creating your initial idea, and then being presented with a range of images, varying slightly in styles.
From there, users can select which direction they’d like to go in, and which variation is their favourite. Upon choosing that option, they’re then shown an array of options that are more tuned to that variation. This process can then be repeated numerous times until we finally reach the desired outcome.
What does this mean for the future of art?
AI art is a clever concept that’s still in its early stages. So, it’ll be interesting to see how it progresses in the future, and how results can improve over time. For now, I think artists and designers are safe, but if the terminator movies have taught me anything… it’s only a matter of time until the machines take over!
Want to try it yourself? There are a growing number of sites you can use to experiment with AI-generated art, which includes:
Midjourney is a discord-based program which offers great flexibility in what you want to create. When signing up, you’ll need to create a discord account, but midjourney offers a great free trial period. So, you have plenty of time to create images and get used to the software. Just sign up online!
DALL-E is a fantastic program, which creates images accurately based on your requests. There is a waiting list to gain access to the site though, so we’d recommend signing up as soon as possible.
Stable Diffusion is a newly launched platform which is free to use and features great setting sliders to adjust how images are generated. The program is also accompanied by a great beginner’s guide on how to get the best out of the software.
Just try these different programs out for yourself, and tag us in your newly created works of art!