TedJun 28, 2024

I've seen more than a few folks on my Mastodon timeline express enthusiasm about Glaze, a system whose goal is to protect artists against style mimicry by generative models.

If you're using this work, or otherwise have heard of it and thought "wow this is neat I'm happy people are doing it", I strongly recommend you read this blog post by security & privacy researchers Nicholas Carlini & Florian Tramèr about evaluating Glaze's claims: https://spylab.ai/blog/glaze/

Glazing over security | SPY Lab

We discuss the security of the Glaze tool, and how the authors' actions may not be in the best interest of their users.

SPY Lab
Show threaddemofoxJun 30, 2024
@tedted I've had 2 incidents where I tried to point some of these things out to the authors of glaze. The first time, some cs professor involved started hurling personal insults at me, saw my bio and asked "how did someone like you get a job at nvidia?" and then he blocked me. Second time was not much better. It really would be nice if it could work, but as you point out, unfortunately it cannot!
Show threadMaxTheFox

@demofox @tedted @nyrath It's basically the digital equivalent of homeopathy.

Cheap way to fix a problem.
No side effects compared to other measures.
Hyped up by both its creators and true believers.
The proposed mechanism for it makes sense on a surface level.
Yet the only "evidence" that it does literally anything, comes from its inventors!

Jul 1, 2024 at 2:49amWeb