Stable Diffusion is a big deal

Maybe this current explosion in the relevance and visibility of this kind of AI model will finally lead us to rethink how insanely nonsensical our IP systems are. I'm not holding my breath, but there's hope that this sort of thing will (combined with situations like the HBO debacle) clarify the need for massive IP reform in the cultural zeitgeist.

The problem here isn't that the model was trained on copyrighted works, the problem is copyright itself and a cultural focus on collecting rent. We probably need UBI or equivalent to deal with the outcomes here in a healthy way.

Fair use is a defense. You can still be sued for fair use, you have to go to court and prove to a judge or jury that your work constitutes fair use.

After using SD heavily for a week, I half agree with this. It is incredibly disruptive, and it's wild how much it accelerates the creative process. I'll give you that.

But two things I've noticed:

First, artists will still have a massive advantage over non-artists with this tool. A photographer who intimately knows the different lenses and cameras and industry terms will get to a representation of their idea much faster than someone without that experience. Without that depth of knowledge, someone might have to rely instead on random luck to create what's in their head. Art curators might be well-positioned here since having a wide breadth of knowledge and point of reference is their advantage.

Second, we need the ability to persist a design. If I create a character using SD, I need to be able to persist that character across different scenarios, poses, emotions, lighting, etc. Based on what I know about the methods SD/Midjourney/Dall-E are using, I'm not sure how easy this will be to implement, or if it's even possible at all. There will always be subtle differences and that's where being an artist who can use SD for inspiration instead of merely creation will retain their advantage over a non-artist.

That said, holy crap. This tech is insane.

There are really significant, novel copyright issues implicated by these large generative models trained on other people’s IP.

If you take a step back, you can see that there are different ways to frame what is happening. One frame is: “Defendant built an algorithm that memorized features of Plaintiff’s IP. Defendant’s algorithm recombines parts of those features in order to produce works in the same domain that compete with Plaintiff’s work, all without Plaintiff’s consent.”

Bear in mind that copyright holders are among the most litigious out there. If generative art becomes as big a deal as some people expect, they will have every incentive to use their huge litigation budgets to claim a piece of the action.

Because the law develops very slowly, the legal process has not yet had the occasion to really evaluate what transformative use means in this novel context. I’m personally interested in seeing where things go, but it’s going to be a while before we know where the law is headed.

What about using this tech for ideation and artists for production?

You could use Stable Diffusion et al to create new characters based on a prompt, then farm the concept out to artists to produce individual works. Kind of like hiring a super expensive agency to design your new logo or brand identity, then using a stable of in-house designers to translate the concept into UI, ads, etc.