*Amore Infernale*: The AI Illustration Contest
$10,000 in prizes, sponsored by Steve Kuhn
Sixteen years ago, I wrote and storyboarded my first graphic novel, Amore Infernale. Based on my original role-playing game of the same name, Amore Infernale is the story of three superheroes solving an opera-themed mystery in Verona, Italy in 2005. Vague spoiler: By the end, they discover that Shakespeare got the story of Romeo and Juliet very wrong!
Over the years, I’ve tried to find an artist to actually draw my book, but nothing worked out. So in 2020 during high Covid, I ran an illustration contest on Freelancer.com. The winner, Aljon Dave, is tremendously talented. He completed the first of four chapters. Check out how awesome some of his pages are…
Unfortunately, Dave subsequently resigned. I thought about hiring someone else, but I loved his style so much that I didn’t have the heart to restart the whole project… until the last time I was hanging out with my dear friend - and founder of Major League Pickleball - Steve Kuhn.
Steve is an AI enthusiast - and at this point, so am I. He’s also a big graphic novel fan. He bought a thousand copies of my Open Borders. Which piqued his interest in the following question:
Is AI ready to draw a full graphic novel - or at least greatly speed up a traditional artist’s work?
To unlock the answer, Steve offered to fund a new AI illustration contest for Amore Infernale. Subject to two key conditions:
Artists are not only allowed, but strongly encouraged, to take full advantage of any AI they wish.
Artists must share details on exactly how they used AI. The software, the prompts, and so on.
After wondering how I could be blessed with friends like Steve, I accepted. I just announced the new contest this morning. Everyone is welcome to try their hand. Here’s the full write-up:
I am Bryan Caplan, Professor of Economics at George Mason University, and the author of the New York Times Bestselling non-fiction graphic novel *Open Borders: The Science and Ethics of Immigration*. For confirmation of my identity, see
https://twitter.com/bryan_caplan/status/1681026328290824193.
As a side venture, I've written *Amore Infernale*, an entirely fictional graphic novel. You can download the full script and storyboards here: http://bcaplan.com/ai.pdf. For a previous artist’s attempt at chapter 1, see: http://bcaplan.com/aidave.pdf.
Your mission:
Combine your artistic ability with any Artificial Intelligence (AI) software to beautifully illustrate my book.
Compensation:
With the help of generous benefactor Steve Kuhn, I have $10,000 to fund this entire project. I plan to run this contest in THREE stages, each of which releases some of this prize money, as I explain below.
Conditions:
Any combination of human and AI output is welcome. Entrants can be professional artists who know a little AI, AI professionals who know a little art, or anyone in between.
Crucially, when you submit your final drawings, you also need to explain *precisely* how you produced the final product. What AI program(s) did you use? What were your main prompts? How did you get from AI pages to the finished product?
My Artistic Vision:
I like stunning representational art. In the attached files, you can see some of my previous favorite attempts to illustrate *Amore Infernale*. I also favor traditional color schemes, the kind you'll see in Tintin or P. Craig Russell's *Ring of the Nibelung* adaptation. Even more importantly, though, I love art with a passionate vision. If you're ever in artistic doubt, remember that my favorite composer is Richard Wagner!
The Contest:
The first stage of the contest will be open to all. I will award ONE first prize of $1000, plus four runner-up prizes of $250 each. During the first stage, draw any 10 pages of my book to show off your craft, vision, determination, and attention to detail. Be sure to explain how you are blending human skill and AI to produce the final pages. The first stage will run for 30 days.
The second stage of the contest will be open only to the five semi-finalists, plus no more than two others that I personally approve. During the second stage, the goal is to BEAUTIFULLY ILLUSTRATE AS MUCH OF THE BOOK AS POSSIBLE. I will award ONE final prize of $2,000 to whoever does the best overall job. The second stage will also run for 30 days.
The third stage: The winner of the second-stage gets $6,000 when they finish the book to my satisfaction. If the winner does not want to finish the book, or makes little progress over a long period, I retain the right to reassign this opportunity to one of the other contestants in the second stage.
If you look at my previous contests, I am diligent at providing artists with careful and honest feedback. Whether you are in striking distance of winning - or don't stand a chance - I will let you know. I know contests are stressful, but I don't want anyone to feel mistreated.
While I own the final copyright and all royalties, the final contest winner will receive full artistic credit on the cover and in any promotions I conduct. Whoever you are, I would sincerely love to make you famous!
If the contest goes well, I hope to publish the finished book commercially. If that doesn’t take, I’ll figure out another release route. False modesty aside, I think my fiction is good enough for Hollywood, back when Hollywood was good. Long-time readers will probably be able to tell I wrote Amore Infernale, but there’s no didactic agenda. Just emotional truth, and art for art’s sake.
In comic art there are some artists who use a lot of photo references, to the extent that a panel looks like a collage of photos that they traced and added stuff to. Greg Land is a famous example. Their work tends to have an Uncanny valley look. They do work faster than traditional artists though.
I suspect that AI is not at a point where it can draw a comic by itself, but it can generate images that an artist can combine and trace over to form a complete comic. This might be able to speed up the work of artists considerably, while eliminating much of the Uncanny valley traced photo look.
Sounds fun and interesting; I for one would be really interested in how much one can currently push AI art to create a complete work, and not just one-off pictures. Many thanks to Steve for funding it!
As an aside, this comes at a slightly awkward time as the new much-improved version of Stable Diffusion (which one pretty much has to use to craft a work of comic-length consistency) will release this month.
After that is released, I expect there to be some weeks before all the tooling (ControlNets, checkpoints...) are updated to support the new model. This means there might be a really short window to submit to the first stage of the contest if you want "current gen" quality!