How AI Forgers Bypass Style Protections to Mimic Artists' Work

Table of Links

Abstract and 1. Introduction

2. Background and Related Work

3. Threat Model

4. Robust Style Mimicry

5. Experimental Setup

6. Results

   6.1 Main Findings: All Protections are Easily Circumvented

   6.2 Analysis

7. Discussion and Broader Impact, Acknowledgements, and References

A. Detailed Art Examples

B. Robust Mimicry Generations

C. Detailed Results

D. Differences with Glaze Finetuning

E. Findings on Glaze 2.0

F. Findings on Mist v2

G. Methods for Style Mimicry

H. Existing Style Mimicry Protections

I. Robust Mimicry Methods

J. Experimental Setup

K. User Study

L. Compute Resources

3 Threat Model

The goal of style mimicry is to produce images, of some chosen content, that mimic the style of a targeted artist. Since artistic style is challenging to formalize or quantify, we refrain from doing so and define a mimicry attempt as successful if it generates new images that a human observer would qualify as possessing the artist’s style.

We assume two parties, the artist who places art online (e.g., in their portfolio), and a forger who performs style mimicry using these images. The challenge for the forger is that the artist first protects their original art collection before releasing it online, using a state-of-the-art protection tool such as Glaze, Mist or Anti-DreamBooth. We make the conservative assumption that all the artist’s images available online are protected. If a mimicry method succeeds in this setting, we call it robust.

In this work, we consider style forgers who finetune a text-to-image model on an artist’s images—the most successful style mimicry method to date (Shan et al., 2023a). Specifically, the forger finetunes a pretrained model f on protected images X from the artist to obtain a finetuned model ˆf. The forger has full control over the protected images and finetuning process, and can arbitrarily modify to maximize the mimicry success. Our robust mimicry methods combine a number of “off-the-shelf” manipulations that allow even low-skilled parties to bypass existing style mimicry protections. In fact, our most successful methods require only black-box access to a finetuning API for the model f, and could thus also be applied to proprietary text-to-image models that expose such an interface.