The AI art that most terrified and amused us over the past year.
Of all the things that 2022 may be remembered for, for creatives it will be the year that generative AI art exploded. In the blink of an eye, AI art generators went from being an amusing curiosity to a powerful technology that's available to everyone and has become impossible for creatives to ignore.
The advances in text-to-image AI art generation have been dramatic, resulting in models that can generate more realistic images and more reliable results super fast. The speed and ease of use means that users can let their imaginations run wild, creating imagery by simply typing what they want to see. And that's led to some weird and wonderful creations.
If you missed the fuss and you’re still not sure how the tech works, see our piece on how to use DALL-E 2. Very basically, AI image generators have been trained on millions of combinations of existing images and captions so that they can produce new images based on text prompts. The AI doesn’t really “know” what it’s doing, it only knows what kind of imagery certain descriptions are associated with. And this can lead to some fascinating results. Here are three of the examples that inspired the good people of internet this past year.
Virtua Fighter gets a makeover
Virtua Fighter's graphics get a update in this AI art project (Image credit: Sega / Colin Williamson via Twitter)
One of the most fascinating uses of AI art generators is to answer all those 'what if...?' questions that can come to us. What if Apple designed a toaster? What every movie starred Nicolas Cage? What if the classic 90s fighting game Virtua Fighter was remade with modern graphics? We might normally dismiss such ponderings without too much thought, but with AI art generators we can now satisfy our curiosity.
Tech artist Colin Williamson used text-to-image tech to find out to enhance what 30 years ago was the cutting edge in video game graphics, feeding characters from Sega's Virtua Fighter through the AI image generator Stable Diffusion to turn the 3D polygon graphics into photorealistic images (OK, kind of). The results were both impressive and quite amusing.
Pixar's John Wick
I typed (John Wick pixar poster) in #dalle ........ Life is weird 🤣 pic.twitter.com/LSAdIbuEkiSeptember 19, 2022
Here's another 'what if...'. What if Pixar took over the action film franchise John Wick turning the story of the former assassin out to avenge the death of a beloved puppy into a family-friendly adventure? AI art generators can give us an idea. The artist BossLogic says he typed 'John Wick Pixar Poster’ into the DALL-E 2 to get the image above, and, well, it does actually make us want to see this film.
As is often the case with AI, the faces are slightly terrifying, with one of Wick's eyes apparently melting. Pixar also seems to have brought Wick's lost puppy Daisy back from the grave. "John Wick if Tim Burton directed," one person suggested on Instagram, while someone thinks it "feels like the cover of a Little Golden Book". Suffice it to say Pixar has no involvement in the upcoming John Wick Chapter 4, but the film does have an awesome optical illusion poster.
Get the Creative Bloq Newsletter
Daily design news, reviews, how-tos and more, as picked by the editors.
Loab
Don't worry, this AI art creation doesn't really exist (Image credit: Supercomposite)
But weird AI art isn't all unlikely mashups and radical makeovers. AI art generators can also be used to create entirely new monstrosities. Loab is an imagined woman painted by an AI and she's been haunting the internet.
The multimedia artist and 'Loabmancer' Supercomposite accidently created Loab when tinkering with an AI art generator. The negative prompt 'Brando::-1' to try to generate 'the opposite of Marlon Brando' generated a logo with the letters "DIGITA PNTICS", so the artist then tried using these letters as a negative prompt to see if it would generate an image of Brando. It didn't. It created Loab, a kind of digital demon that people are describing as AI art's first cryptid (a beast often claimed to exist but never proven).
So where does this AI monster come from? She's presumably a composite from many images the AI was trained on, but things get weirder. When Loab is fed back into the AI to create a create variations, the results get even more macabre, with Loab surrounded by bloody body parts, severed heads and creepy dolls.
Loab shows that a lot of the magic, and horror, of AI art at the moment comes from experimentation and accidental discoveries. Creatives are still trying to work out what the tech can do – and the tech itself continues to take leaps forward almost every month. To learn more see this comparison of the best AI generators and our pick of other weird AI art that's been doing the rounds online.
Joe is a regular freelance journalist and editor at Creative Bloq. He writes news, features and buying guides and keeps track of the best equipment and software for creatives, from video editing programs to monitors and accessories. A veteran news writer and photographer, he now works as a project manager at the London and Buenos Aires-based design, production and branding agency Hermana Creatives. There he manages a team of designers, photographers and video editors who specialise in producing visual content and design assets for the hospitality sector. He also dances Argentine tango.
