DEV Community

Cover image for Art & AI
Jason Hornet
Jason Hornet

Posted on • Updated on

Art & AI

(Cover made with Dall-E 2)

For some time now, images generated by artificial intelligence have spread across the internet and artistic events. Maybe you've played with tools like this yourself. Stable Diffusion is the most popular example today. But there are others, like Dall-E 2 and Midjourney.

Many results are so interesting that they end up in communities like DeviantArt, ArtStation and Pixiv. In these, there are no restrictions on art generated by artificial intelligence.

Platforms like NewgroundsInkblot Art and Fur Affinity have implemented policies that ban or restrict the reach of images made by artificial intelligence, says Andy Baio, from the site Waxy.

For example, in the announcement about their new policy, Fur Affinity justified the decision to ban computer-generated images:

“Our goal is to support artists and their content. We don’t believe it’s in our community’s best interests to allow AI generated content on the site”

The problem is that engines like Midjourney and Stable Diffusion have become so popular that art platforms are being flooded with AI-generated content.

AI generated images (mosaic from Midjourney)AI generated images (mosaic from Midjourney)

In the end, this seems to be a problem of quantity, not just artistic merit.

While artists take hours to produce their works, artificial intelligence can do it in a matter of minutes or seconds.

A good example of this is the controversy in the work Théâtre D’opéra Spatial made by Jason Allen using Midjourney, his work won the first place against 18 competitors, thus winning the prize of US$ 300.00.

Several debates have arisen after Allen’s case was exposed. The winner responded to his critics saying he knew the matter would be "controversial.”

“How interesting to see all these people on Twitter who are against AI generated art are the first to throw a human under the bus by discrediting the human element! Does that sound hypocritical to you?”

says Allen.

Allen, who is president of a Colorado-based tabletop game company called Incarnate Games, said he believes his work is an extension of his own creativity. After all, he defines the parameters in which the AI creates and after that he processes them in Photoshop and finishes in Gigapixel.

“I am exploring a special prompt that I will post later, I created hundreds of images using it, and after many weeks of fine-tuning and curating my gens, I chose my top 3 and print them on screen”

he wrote in a post before the winners were announced.

Despite all the controversy being generated through this discussion, some of these images can be really interesting. Therefore, community administrators are faced with a dilemma: release, release with restrictions or ban it.

Whatever is the decision, the discussion on this matter is far from over.

Top comments (5)

Collapse
 
leob profile image
leob • Edited

So they want to forbid AI-generated content - how, exactly? Seems a lost fight even before it started.

That 'discredited' winner makes a valid point - you can argue that it's just another tool in the toolbox ... even without AI most of this stuff is "generated" using drawing software, nobody's putting the individual pixels there. You want "real" and unaided art that cannot be faked or messed with, then go back to the physical atelier (art studio) where people put real paint on a real canvas.

This tells me that the whole 'digital art' thing is being taken a bit too seriously, with official competitions and prizes and all that - time to take a few steps back and just see it as a laudable but glorified hobby. "Real" artists are found in art workshops, putting real paint or ink on real canvas or paper.

P.S. looking at the images in this post, I honestly like none of it - all of it looks perfect in a technical sense, but the level of artistry seems zero, I just see platitudes and banality, everything seems 'science fiction' or "fantasy" themed, hundred percent cliché-for-the-masses ... just not my thing.

Collapse
 
mellen profile image
Matt Ellen

"A tool in the tool box" isn't really a good analogy. The AI has been trained on thousands of different artists. It can be asked to generate an artists previous work, and will. It's not like Gausian blur or a mozaic effect. It's practially a breach of copyright.

There isn't a good analogy for what is happening here becaue the scale is impossible for humans to achieve without the AI. This isn't a mechanical hammer, this is Thomas Edison passing off the light bulb as his own idea.

Yes, prompt development is a skill, but compared to the set of training data, it's nothing.

Collapse
 
leob profile image
leob

The point is it might be undetectable, so what can you do ... people will do anything they want, this stuff can't be policed.

Collapse
 
wadecodez profile image
Wade Zimmerman

AI art is cool but it destroys an entire economy. Now anyone can spend like 10 bucks to create the piece of their dreams.

Meanwhile, people making use of AI can make a ton of money for practically zero effort.

Reminds me of elephant paintings for some reason.

Also what is it with humans and enslaving life for cheap entertainment?

Collapse
 
integerman profile image
Matt Eland

This is really interesting. I think I'm going to use that anecdote of the art competition in an upcoming conference talk on the impact of AI. Thanks for sharing this!