Leveraging Optimization for Adaptive Attacks on Image Watermarks
This work addresses the security of watermarking for image generators, highlighting vulnerabilities that could undermine misuse prevention, but it is incremental as it focuses on improving attack methods rather than proposing new defenses.
The paper tackles the problem of assessing the robustness of image watermarking algorithms by designing adaptive attacks as an optimization problem, resulting in attacks that break all five surveyed watermarking methods with detection accuracy reduced to 6.3% or less at no visible quality degradation and requiring less than 1 GPU hour.
Untrustworthy users can misuse image generators to synthesize high-quality deepfakes and engage in unethical activities. Watermarking deters misuse by marking generated content with a hidden message, enabling its detection using a secret watermarking key. A core security property of watermarking is robustness, which states that an attacker can only evade detection by substantially degrading image quality. Assessing robustness requires designing an adaptive attack for the specific watermarking algorithm. When evaluating watermarking algorithms and their (adaptive) attacks, it is challenging to determine whether an adaptive attack is optimal, i.e., the best possible attack. We solve this problem by defining an objective function and then approach adaptive attacks as an optimization problem. The core idea of our adaptive attacks is to replicate secret watermarking keys locally by creating surrogate keys that are differentiable and can be used to optimize the attack's parameters. We demonstrate for Stable Diffusion models that such an attacker can break all five surveyed watermarking methods at no visible degradation in image quality. Optimizing our attacks is efficient and requires less than 1 GPU hour to reduce the detection accuracy to 6.3% or less. Our findings emphasize the need for more rigorous robustness testing against adaptive, learnable attackers.