SOURCE-LINKED INTELLIGENCE

One Prompt Is Enough: Watermark Laundering Through Foundation Image Models

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Invisible watermarks are typically evaluated against predefined perturbations such as compression, blur, noise, cropping, and denoising. Public foundation image models expose a distinct threat: an attacker can submit a watermarked image with a single reconstruction prompt and obtain a visually faithful output from which the invisible watermark can no longer be decoded reliably. We formalize this failure mode as watermark laundering and evaluate it using a joint payload-fidelity profile that combines bit error rate (BER) with visual and semantic preservation. Across six OpenAI and Google image

Read original source ↗ Open in workspace

recordType

paper

region

Global