Topic
watermarking
Technology Coders Say They Already Found Workarounds to Claude’s Invisible Watermarks
Developer Guillaume Meyer published a code override removing Claude's invisible watermarks within hours of Anthropic's announcement, according to WIRED. The bypass, which rewrites text with non-watermarking LLMs, raises compliance questions for enterprises under the EU AI Act, which threatens fines up to 3% of annual turnover.
From Construction to Injection: Edit-Based Fingerprints for Large Language Models
A new arXiv paper introduces an end-to-end injected fingerprinting framework for large language models (LLMs), addressing the dual challenges of imperceptibility and robustness. The proposed methods—Code-mixing Fingerprints (CF) and Multi-Candidate Editing (MCEdit)—aim to provide reliable ownership verification in black-box deployments without degrading model utility.
SAMark Watermarking Breaks Paraphrase Robustness Barrier for AI-Generated Text
Researchers propose SAMark, a self-anchored text watermarking framework that achieves up to 90.2% true positive rate under paragraph-level paraphrasing attacks, outperforming the strongest prior baseline by more than 30% on average. The method breaks the robustness-quality trade-off by using multi-channel hyperbolic scoring and diversity-aware filtering.