AI Images as References for Live Artists
So, I’ve written a few articles on my exploration of human-prompted, AI-generated images. There is going to be more and more public discussion about this, and I think articles like this one at Kotaku, are important to read and consider. Of course, such articles are generally an undifferentiated mix of fact and opinion, but even where I disagree with the opinions I want to be open to them and see if any new points of view or data are presented that may alter my current position on the issues.
I also want to keep exploring various possible ways to use human-prompted, AI-generated images. One of the things I’ve mentioned is the idea of using such an image as a visual reference for an artist.
So, I did.
Here is an image of a lich I got from prompts fed into MidJounrey.
I really like this image, including it’s sartorial style, but there are (at least currently) significant limitations I’d have to contend with if I wanted to use this commercially. Not the least of those is I have no way of creating different images of the same being.
So, what if I gave this as a reference to an artist I like? (And, you know, pay them to create more art.)
So, I took this to Jacob Blackmon and asked if he was willing to participate in my experiment (at his standard rate). He was, and sent me this sketch:
This highlights two of the things I love about working with Jacob. First, he did a LOT of design work in this piece, flowing from the reference image, but absolutely building well beyond either it or even standard fantasy lich images. Secondly, his sketch stage has enough detail for me to see where he is going and give useful feedback or ask for alterations. In this case, I gave an immediate thumb’s up.
So, that brought us to this, Jacob’s final.
This is awesome, it shows Jacob distinct and developing style, I love it, i can use it, Jacob got paid for it, and it would not have existed without MidJourney returning the top image in response to prompts I gave it. It’s not a duplicate of the original (nor did I ask for it to be), but it clearly uses that image as the umping off point for a new design.
I like this result, and can easily see going this route again. It remains to be seen how issues of legality, ethics, and public opinion shake out on any use of human-prompted, AI-generated images, but I found this a useful project to help me explore my own thoughts on the subject.
And, in this case, it put money in the pockets of an artist I like, which is always a plus.
Speaking Of Money…
I have a Patreon. It helps me carve out the time needed to create these blog posts, and is a great way to let me know what kind of content you enjoy. If you’d like to see more essays on industry issues, or Pathfinder 1st or 2nd edition, 5e, or Starfinder content (or more rules for other game systems, fiction, storytime posts, game design articles, worldbuilding tips, whatever!), try joining for just a few bucks and month and letting me know!
If you prefer, you can drop a cup of support in my Ko-Fi. It’s like buying me a cup of coffee, but more convenient!