AI Art IS Art

I don’t know about you, but lately my social feed has been absolutely crammed to the gills with folks testing out the latest AI (artificial intelligence) tools. From crazy conversations with ChatGPT, to generating awesome generative art with Stable Diffusion or DALLE or Midjourney, to creating songs with Soundraw, and on and on and on.

It seems that generative AI has just hit the crazy acceleration point of an exponential curve, and it’s pretty impressive across the board.

But not everyone is happy about it.

For example, check out some of the awful replies in this Twitter thread:

The people in this thread, apparently largely artists, seem to be some pretty fucking douche people treating a stranger like garbage.

Here’s a guy, playing around with some new AI tools, trying to be creative, and holy shit these butthurt artists are absolutely losing their shit and attacking this guy like he’s the antichrist…wild.

And then yesterday and today this image starts making the round via artists:

So, what the hell is going on?

It’s pretty simple actually…artists are scared shitless. 😂

You see, generative AI tools like Stable Diffusion and DALL-E 2 have gotten so good, so fast, that already “starving artists” are seeing what they perceive to be an existential threat to their livelihoods looming over them.

There are some folks pushing back on the other elements (text generation, sound generation, etc.), but most of the backlash is art-centric.

Ironically, these clearly fearful artists are spending a LOT of time pointing out how absolutely terrible these tools are, how they generate nothing but shit, how none of it is really art, yada yada yada.

Well…

AI-Generated Art Wins Art Competition

Sure, the tools aren’t perfect. They suck at hands, and irises, and sometimes heads, and lots of other little things…but they are LITTLE things, and these tools are improving wicked fast. Any flaws they have now will be ironed out in 6-12 months if that. What then?

Here’s the thing though…these artists, fearing that their value to the world is about to be eaten by machines, are wildly overreacting.

That isn’t to say that none of these tools will negatively impact any artists, I’m sure they will, but the degree to which they will harm vs. help I believe depends entirely on how these artists approach using these tools.

Instead of looking at these tools with fear, with a zero-sum mindset, why can’t they see the incredible opportunity these offer?

In the AI world, there’s a term, Centaur, that refers to an AI and a human working in tandem. And, time and time again, researchers have found that AI + Human, beats AI or Human alone EVERY. DAMN. TIME.

There’s a LOT to art that can be slow, tedious, monotonous, and repetitive. Some artists may not think so, but some sure do.

Why not, as an artist, take a look at your workflow and see where these tools could shave off time? Or see if there are areas where the AI is actually better than you and can augment what you create or even teach you something? What about using it to help generate new ideas, or to modify existing ideas, or to mash together things like spit balling concepts?

Here’s a detailed post, from an artist, on how they’re integrating AI tools into their workflow, and its’s awesome!

If you’re doing art in some capacity as a career, and are being paid by the piece, that means that the single most important thing you could optimize in your workflow is income per hour worked. If you used to make $500 for a commission and it took you 20 hours, but using an AI tool coupled with your expertise can cut that down to 5 hours, you just went from making $25/hr to $100/hr.

Who the fuck doesn’t want to 4X their income per hour worked???

And this is just the most obvious opportunity in my eyes, I’m sure there are many others!

Are there some art projects that won’t happen anymore because it’s cheap and easy to do it with AI?

Probably.

Will these tools be nothing but sunshine and rainbows?

No, because humans are often idiots, and all technology is a double-edged sword.

But are they a net good for humanity?

Without any doubt in my mind, because they help more people to fan their creative spark.

Back to art, there will surely be people who put a premium on human art, and there will be new projects in other areas, and if you learn to work with these tools instead of fearing them perhaps you’ll end up a far, far better and more successful artist than you otherwise might have been?

In addition, beyond the existing artists and how these tools could enhance their workflows, I see a much more important aspect to generative AI tools.

They will, without any doubt in my mind, unlock much more human creativity than most could possibly imagine. Kevin Kelly agrees.

Why?

Because they lower the barriers to entry.

There are many times many people out there who feel a creative spark, but for one reason or another have never pursued it.

Maybe they lacked opportunities. Maybe they wanted (or simply had to) pursue more practical work. Maybe they felt like they sucked, and nothing they ever made matched what they were envisioning. Maybe they have to work 3 jobs to support a family and don’t have time to work on art or improve their skill?

So, so many reasons why someone might not be what gatekeeping artists consider a “real artist”, but they wish they could be.

Well, now they can.

With generative AI tools they now have the ability to feed that creative spark, to make art even though they lack the time, tools, or skill.

Does throwing a few words into an AI tool make one an artists? No, of course not. It’s the degree to which you get your hands dirty, your vision, and your process.

But some folks are using these tools as just one step in a rather elaborate and creative process to generate images that match a clear vision, modifying them in Photoshop, blending and layering elements from different sources, testing and tweaking and refining prompts, settings, and on and on and on.

And sure, the hoity toity judgmental art crowd might turn their noses up at it, say it’s not art, revile those who use these tools…

Fuck ’em.

Gatekeeping is for fearful cowards.

Creativity, and more importantly CURIOSITY, should be encouraged, not reviled.

And, let’s be honest…art is, as ever, SUBJECTIVE.

What one calls art, another calls shit, and vice versa. It has ever been thus, and will ever be.

Subjectivity is part of what makes art ART…it makes you think, makes you feel, and it does much of that completely independent of the artist and their vision or intentions.

If someone thinks something is art, it IS art. Pure and simple.

If people will pay $120,000 for a banana taped to a wall, or $69,000,000 for a Beeple JPG, you know what? Nobody can say what is or isn’t art, only what is or isn’t art *to them*.

Moving on, the next great gripe from the art tarts is that these tools are somehow stealing, plagiarism, copyright infringement, etc. etc. whine whine whine.

Alas, once again, they are categorically wrong. Let’s count the ways:

  • It is not possible to protect “style” via IP laws
  • Many, many places online where these images used to train these AI tools are pulled from include rights to do stuff like this in the terms and conditions…you did read those, right?
  • The next iteration of these tools will generate even better results using only a small number of fully licensed images
  • At least some of these tools already fully licensed all the images they used to train these datasets
  • Current IP laws protect the right to fair use, and to create derivative works based on existing art, which these art tools are compliant with (as far as I can tell, and I have a decent amount of experience with visual IP law from my time at Getty Images)
  • Many of these tools grant the person using them a legal license to use these images as they see fit, because they are considered unique works
  • If you go to art school, guess what you’re trained on? Other people’s works, other styles, etc. If someone says “Hmmm, I want to learn to paint in the style of [insert artist name]” they are 100% free to look up all of their works, learn that style, and create whatever they see fit in that style. There are no rules that say you can’t use a computer to do the same thing.
  • Art includes mixed media, using other tools and mediums to make new art…AI generated art is just the newest tool, the newest medium. It was painting robots before that.
  • Nothing is 100% original, everything is at least some % derivative (built on top of or inspired by other work), that’s just the nature of the beast

Repeat after me:

EVERYTHING IS A REMIX

If you’ve never seen this video series, it’s 100% worth a watch 😎

From what I can see, artists bitching about AI are just the newest rendition of the famous Luddite Weavers…and their bitching will be no more effective than it was for the Luddites of old.

Technology will continue to advance. Tools like these will get better and better. More new things will be possible.

This will result in some jobs changing, and some jobs going away, as always. Remember, ~100 years ago the largest portion of folks worked on farms. The times, they are a changing.

So, to sum it all up:

  • AI art IS art
  • AI tools can be used my *smart* artists to augment their skills, speed up their processes, and earn more for their time
  • AI tools help to unlock more creativity in more people, which is great for those people and without a doubt a net-positive for society
  • AI image generation tools are, or at least appear to be, legally compliant (pretty sure courts will uphold this)
  • Everything is a remix

So, to all those artists out there bitching about AI, I’ll say this.

Adapt or die.

Evolution rewards the adaptable, and punishes the stagnant. Take a look back through time and see how often resisting a new technology worked out for those resisting.

Whining will change nothing, and you’re just going to end up looking like assholes, and like so many before you end up on the wrong side of history.

ADAPT. OR. DIE.

 
 
There Is No Underpopulation Crisis

Sam McRoberts

Author of Screw the Zoo. CEO of VUDU Marketing.