Context

Context: I typed this up with a bit of anger in a Discord during an AI Art conversation when it swerved to “but soon most to all art will be AI anyways,” and I think I finally figured out what I want to say about these kinds of conversations.

I added some extra stuff at the end, of course.

Do You Even AI, Bro?

So what some people seem to be saying is that board games, like all art, will become purely designed by AI, and thus as designers and artists we should stop doing this and focus on producing value for company bottom lines and only consume.

I feel like this is always the logical endpoint of these conversations.

And like… has anyone in these conversations ever written AI systems?

Do you know how hard it is to get AI systems to create new things rather than rehash old ones?

This is always the thing I don’t understand about these conversations. AI cannot do right now what people are proposing that it will do.

This is kind of like saying photographers don’t have to do anything but point and shoot. Any pro photographer who has spent days trying to catch the golden hour and then process photos to actually look good will laugh in the face of such a concept.

And in a way, these conversations are both harmless (they will change nothing about what AI can or can’t do) and harmful (they fool people into believing more of AI than is possible or even wise).

History Is a Flat Circle

I remember when CGI first came on the scene in movie making. Suddenly folks were saying, why, any movie can make anything happen with CGI on any budget whatsoever! Who needs set designers, prop designers, makeup artists, costume designers, or stunt doubles?

And now look where we are. People don’t like a lot of CGI because it too often hits an uncanny valley of physics – not because “it’s not done by a human.”

Even if AI becomes more capable in the next few decades – that’s the story of how it always goes. Skill and talent in using it will matter, and it will not take the place of artists.

Heck. Good AI art still relies heavily on a human having the artistic ability to guide it.

A camera can never just point and shoot and get you art gallery level results.

No film camera ever existed that could create a masterpiece without someone who understood the art of filmmaking.

And no AI tool will be able to make any of that possible.

What Would It Take?

Not until we crack how the brain even begins to work.

We are not there yet. We have never been there. We will not be there in a century. We can’t even fathom how incredibly deep brains go, we don’t even know what things we need to know yet.

I think that’s the synthesis of what I’ve wanted to say about these kinds of conversations for a very long time.

And I am tired.

The Impenetrable Future

We can’t predict precisely what’s going to happen in the future with AI (generative or other), any more than we could with other new technologies from the past.

For instance, the end result of CGI and VFX wasn’t a sudden freeing of labor in the movie industry to pursue higher goals – the end result was VFX sweat shops. The end result was crunching an entire generation of computer animators out of the industry.

And also lots and lots of bad CGI and VFX. Sturgeon’s Law just can’t be countermanded, it turns out.

New tech is always a genie. You can’t stuff it back in the bottle. And it doesn’t ever grant you the wish you originally wanted.