Screencap from Midjourney’s community showcase
I’ve never seen a technology that has moved as fast as AI Generative Art has in 2022. It’s astounding how much has happened in just the second half of the year alone.
The following is not a comprehensive list, but some of the major highlights.
Open AI’s DALL-E 2
Open AI’s DALL-E2 image (release)
While Open AI’s initial DALL-E was released in 2021, the much anticipated DALL-E2 was released in beta on July 20th to roughly 200 people (artists, researchers, and trusted users) and invitations were sent to 1 million waitlisted individuals.
The craze for access was overwhelming, and DALL-E2 was officially released to the public on September 28th. Over 1.5 million users now generate about two million images per day using the tool.
DALL·E Mini, a prior release of DALL-E was rebranded as Craiyon, and has been available for free here.
2022 Popular AI Art Generators
While everyone was waiting for DALL-E2 access, current AI generators increased in popularity. Some notable AI generators listed below were released prior to 2022, and had a niche following of experimental artists.
Artbreeder — Artbreeder formally known as Ganbreeder, is a collaborative AI-based website that allows users to generate and modify their portrait and landscape images. It originally launched in 2018.
NightCafe — NightCafe Creator — originally a neural style transfer app — launched on Reddit in November 2019. Nightcafe now has a mobile app and the ability to order prints of the artwork.
Dream — Wombo’s Dream is the very first AI art tool that I used. You can generate NFTs from the art, and Wombo is currently releasing DreamBooth, a portrait creation tool.
Jasper.ai Art — Jasper is known for the popular AI writer, but has recently released an impressive AI art generator in the suite of Jasper tools. Jasper has a committed user base. I would recommend you follow
The Jasper Whisperer
for more info on Jasper art.
The Rise Of MidJourney
Midjourney is an independent research lab that produces a proprietary artificial intelligence program. Because it is still considered to be an open beta, Midjourney runs primarily on discord.
Midjourney entered open beta on July 12, 2022. April 2022 — Version 2 July 2022 — Version 3 November 2022 — Version 4
It would be an understatement to say how important and influential Midjourney has been in the AI art space. There are over 4 million Midjourney discord members — a very passionate community.
The Tech Giants And Their Research Labs
2022 was the year when the tech giants announced and demo’d new generative art tools soon to be released.
Google Brain announced two different types of text-to-image models called Imagen and Parti. On Nov 2nd, Google stated that it will be adding Imagen — in a very limited form — to its AI Test Kitchen app as a way to collect early feedback on the technology.
Meta announced their version of an AI image generator ‘Make-A-Scene’. A notable highlight is the ability to start with rough sketches when generating art.
Nvidia has been releasing demos of AI projects and Microsoft is planning on including DALL-E2 in it’s designer app.
While the tech giants are still working on releasing public versions, a new company has emerged in the ranks. Stability AI
Stabe Diffusion and Rise Of Open Source AI art
Stability AI — website landing page
With no doubt, the most important event in the world of generative AI in 2022 happened on August 22nd. That was when Stable Diffusion was released to the public as an open-source model.
The model has been released by a collaboration of Stability AI, CompVis LMU, and Runway with support from EleutherAI and LAION.
Since then, we have seen an insane amount of AI art tools popping up on an almost daily basis. From websites, to google colab notebooks, to plugins, to modules, … any of the free startup tools are likely running on stable diffusion.
A curated list of resources via Stable Diffusion can be found here.
A Year Of Firsts
We have seen 2022 as the year of firsts in AI art. Here are some most notable mentions:
First AI artwork to win an art competition
Jason Allen’s art created on Midjourney titled “Space Opera Theater”
In August, Jason M. Allen submitted an art piece at the Colorado State Fair Art show titled ‘Theatre D’Opera Spatial’ (Space Opera Theater). The piece was created using MidJourney, and it won first place.
First AI generated magazine cover
Cosmopolitan cover created by DALL-E 2
Cosmopolitan editors teamed up with OpenAI and a digital artist named Karen X. Cheng. Using DALL-E 2, they created the first magazine cover that was designed by AI.
First copyright registered AI art
Zarya of the Dawn, an AI assisted Graphic Novel by Kris Kashtanova
Kris Kashtanova, a graphic novel artist who uses Midjourney to create comic books, received a US copyright registration for a comic book titled “Zarya of the Dawn.”
Note: Because of all the publicity and backlash, the US copyright office did reach out to Kris a month later to retract the copyright certification, and I believe that Kris is fighting it at the moment (with the help of Midjourney) — so it is pending.
Final Thoughts
While AI generated art has been with us for a little while (especially in the last decade with amazing new development), 2022 has clearly been the year where AI art exploded in the mainstream.
As AI art tools continue to proliferate into everyday apps and tools that people use, AI art will grow exponentially in 2023. I’m excited to see how AI art advances in 2023.
If I missed anything of importance, which I probably did, please list it in the comments.
If you liked this article, just hit the follow button to see more articles like this in the future.
/w=3840,quality=90,fit=scale-down)
