Watermark Stealing in Large Language Models

arXiv:2402.19361v1 Announce Type: cross
Abstract: LLM watermarking has attracted attention as a promising way to detect AI-generated content, with some works suggesting that current schemes may already be fit for deployment. In this work we dispute this claim, identifying watermark stealing (WS) as a fundamental vulnerability of these schemes. We show that querying the API of the watermarked LLM to approximately reverse-engineer a watermark enables practical spoofing attacks, as suggested in prior work, but also greatly boosts scrubbing attacks, which …

arxiv cs.ai cs.cr cs.lg language language models large language large language models stealing type watermark