logoalt Hacker News

jowsietoday at 6:03 AM1 replyview on HN

The tech doesn't sound that interesting at all. Every AI Degen thread on 4chan and similar has included model fine tuning instructions for a few years now, for the express purpose of cloning an existing artists style. I also find it interesting that they included a quote from an artist pointing out the hypocrisy of using an existing model, trained on unlicensed material, but never actually discussed that particular issue in the article.


Replies

fennecfoxytoday at 10:05 AM

I think you mean LORAs more than a fine tune. Yes, plus there are plenty of online resources to train a LORA as well, CivitAI you can just give it a bunch of images + labels and it just does it for you, the bar is pretty low.