Four of Hollie’s illustrations used to train the AI model (left) and sample AI output (right) -- waxy.org
Los Angeles artist and character designer Hollie Mengert found out that someone used her online portfolio to train an AI text-to-art model. The AI art looks a lot like Mengert's work, at least at a first glance. Andy Baio has a great article about it.
Using 32 of her illustrations, MysteryInc152 fine-tuned Stable Diffusion to recreate Hollie Mengert's style. He then released the checkpoint under an open license for anyone to use. The model uses her name as the identifier for prompts: "illustration of a princess in the forest, holliemengert artstyle," for example.
Artwork by Hollie Mengert (left) vs. images generated with Stable Diffusion DreamBooth in her style (right)
Andy asked Mengert how she felt about having her art used this way without her permission:
"For me, personally, it feels like someone's taking work that I've done, you know, things that I've learned — I've been a working artist since I graduated art school in 2011 — and is using it to create art that that I didn't consent to and didn't give permission for," she said. "I think the biggest thing for me is just that my name is attached to it. Because it's one thing to be like, this is a stylized image creator. Then if people make something weird with it, something that doesn't look like me, then I have some distance from it. But to have my name on it is ultimately very uncomfortable and invasive for me."
Andy also spoke to the person who made the model, Ogbogu Kalu, ("a young Nigerian engineer living and working in Halifax, Canada"):
His take was very practical: he thinks it's legal to train and use, likely to be determined fair use in court, and you can't copyright a style. Even though you can recreate subjects and styles with high fidelity, the original images themselves aren't stored in the Stable Diffusion model, with over 100 terabytes of images used to create a tiny 4 GB model. He also thinks it's inevitable: Adobe is adding generative AI tools to Photoshop, Microsoft is adding an image generator to their design suite. "The technology is here, like we've seen countless times throughout history."
- Canadian prog metal heroes, Voïvod, released a new album earlier this year, Syncro Anarchy. The latest video for the track "Quest for Nothing" may look reminiscent of the band's distinctive art, done by drummer Away (aka Michel Langevin), but it isn't. The video was created by the band's webmaster, Luc Leclerc, using AI art. Michel… READ THE REST
- GhostlyStock generates images that look like stock photos or illustrations: "Freely-usable images summoned from beyond the aether. Energized by AI." The results are quite appalling and I look forward to you seeing more of it. Here's a couple of picks from the prompt "women laughing alone with miter saws": Here's "Newt Gingrich informed that his…
- We thank our sponsor for making this content possible; it is not written by the editorial staff nor does it necessarily reflect its views. As we start to retreat into our holes this winter, activities become less and less exciting (unless you're into puzzles, in which case, you've got some work to do). That said, there's… READ THE REST
- We thank our sponsor for making this content possible; it is not written by the editorial staff nor does it necessarily reflect its views. Mopping your floors can feel daunting, but you need to walk on them eventually! So why not do both at once? You can spice up your cleaning routine as you walk… READ THE REST
- We thank our sponsor for making this content possible; it is not written by the editorial staff nor does it necessarily reflect its views. We all hate waste. Yet you might be surprised at how often a good thing actually goes to waste. For example, have you ever downed a glass of wine, then had…
5ff56afd237d461ab0c3773646985e2a
