Published on
Table of Contents
Side by side: left, a original pencil sketch. Right, someone who sent me an AI 'finished' version derivation of it...
As a webcomic author, releasing my work under free culture license, making my art with only Free Libre and Open Source software, and with popular stories and art, I'm facing a new license problem.
The Creative Commons licenses I used since more than a decade (CC By for Pepper&Carrot, and CC By-Sa for MiniFantasyTheater) miss an option: the possibility to prohibit AI generated derivation.
And I really need now to setup something like that.
AI generated derivation?
I'm speaking here about AI generated images or texts inspired directly by my creation, characters, lore, or stories. That's because nowadays, you can prompt any AI/LLMs for a Pepper&Carrot fan-fiction, and it will just output something really detailed, knowing the characters, their magic, and their way to speak. Everything about my stories, comics and art were trained.
As a consequence, I'm receiving right now daily emails from animation studios, interactive comics publishers, publishers of comics on web or paper, game dev, software dev or improvised AI comic artists asking me for permission of making AI derivation of my art.
This week it is more than three per day.
screenshot: asking to any AI a Pepper&Carrot story is now trivial. Everything was trained.
Hard to manage
Reading all those emails and answering rejection became too exhausting, and with the flow of them, impossible now. Especially since most of them use AI to write long sloppy emails.
I like to imagine this volume of emails could get down if my license clearly indicated a "AI derivation prohibited" after the Creative Commons.
I really wished Creative Commons could come up and update their set of acronyms (BY, SA, NC, ND) to include something like that, but I see no sign of them working on something like that, and when I read their blog, I have no hope they'll ever address this issue.
So, I decided to take initiative and DIY something.
An example of AI derivation: my artwork reused by an account on Temu over a AI generated photo.
Statement
As an artist, I don't want my art, lore, stories, or my comics to have AI generated derivatives. It means derivatives where AI was used to generate or substantially replace the development of ideas, creative process, or visual rendering.
I explicitly here prohibit AI generated derivations based on my art: stories, lore, fan-fiction, script, AI generated image, renderings, and comics.
Examples:
- I don't want a publisher to come with a comic prequel of Pepper&Carrot, AI generated.
- I don't want a Pepper&Carrot video game with AI graphics.
- I don't want a MiniFantasyTheater interactive comic where animations are managed with AI.
This statement will complement and update an already existing "NO AI" contribution policy on all Pepper&Carrot repositories since March 2026.
An example of AI derivation: sent to me by a publisher who wanted to sell on Amazon a prequel derivation of Pepper&Carrot, a proof of concept artwork.
Enforcement
As a European artist, I benefit of being protected by my moral right (see https://en.wikipedia.org/wiki/Moral_rights). It is just required for the author to 'assert' these moral rights before they can be enforced. That's the purpose of this blog post.
To communicate this enforcement, I'll soon write a tiny "AI derivation prohibited" side by side to my Creative Commons licenses, and everywhere I can. I'll start this update from today, and backport it to previous artworks and comics, but it might take time.
Of course, I'm also inviting all my audience to report AI made derivation of my work found on the web or in real life. Some content will be easy to report and get down, some other bigger case might require to go in front of courts, and I think I'm now ready for that thanks to my clear statement.
An example of AI derivation: sent to me by a user who thought I would be happy to receive generated ideas for my weekly comic Avian Intelligence...
Nuances
I'm aware that asking here for a 100% "no AI" derivation is, in the current tech world, pure mission impossible. So here is a bit of nuances.
I'll tolerate the usage of software that received AI contributions. I myself paint and create daily on a Linux operating system, a project that accepted AI code contributions. Same goes for Debian or Ubuntu: the base of my KDE Neon operating system. I don't like their policy, but I accept it as it is because it is not possible right now for me to paint on something else.
I'll also tolerate the use of AI powered tools for grammar, proofreading and translation. Mainly for correcting a human authored text, or to assist in a language one already knows. Over the last years, many web services or tools like Google Translate or DeepL shifted to use AI/LLMs under the hood, the same for many translation software with AI features, and I get the same requests about them over and over. I also want to be clear: I don't encourage AI-assisted translation, or automatic translation. Translation is a human art, and it requires a strong art-direction, a deep cross-cultural knowledge and a certain level of creativity, especially for humor. I'm well placed to know that with Pepper&Carrot, a project with hundreds of translator contributors and 69 languages, 12 years of contributions and community efforts.
Some other nuances might apply, the real goal here of such nuances being to preserve human-made works, while triaging low-effort opportunistic machine generated slops away. I think drawing the line between the two will always be sort of easy.
An example of AI derivation: someone felt the good taste to say they could remake Sintel, the Blender open movies with my concept art and art-direction but in AI, and sent a proof of concept...