Watermark-based Attribution of AI-Generated Images

Several companies have deployed watermark-based detection to identify AI-generated images. However, attribution--the ability to trace back to the user of a generative AI (GenAI) service who created a given AI-generated image--remains largely unexplored despite its growing importance. In this work, we aim to bridge this gap by conducting the first systematic study on watermark-based, user-level attribution of AI-generated images. Our key idea is to assign a unique watermark to each user of the GenAI service and embed this watermark into the AI-generated images created by that user. Attribution is then performed by identifying the user whose watermark best matches the one extracted from the given image. This approach, however, faces a key challenge: How should watermarks be selected for users to maximize attribution performance? To address the challenge, we first theoretically derive lower bounds on detection and attribution performance through rigorous probabilistic analysis for any given set of user watermarks. Then, we select watermarks for users to maximize these lower bounds, thereby optimizing detection and attribution performance. Our theoretical and empirical results show that watermark-based attribution inherits both the accuracy and (non-)robustness properties of the underlying watermark. Specifically, attribution remains highly accurate when the watermarked AI-generated images is either not post-processed or subjected to common post-processing such as JPEG compression, as well as black-box adversarial post-processing with limited query budgets.