Who Gets Flagged? The Pluralistic Evaluation Gap in AI Content Watermarking

TL;DR

This paper highlights the biases in AI content watermarking across languages, cultures, and demographics, proposing new evaluation standards to ensure fairness and robustness before deployment.

Contribution

It introduces three concrete evaluation dimensions for pluralistic watermark benchmarking addressing content diversity and bias.

Findings

Most watermarking benchmarks lack cross-lingual and cultural performance data.

Watermarking fairness standards are lower than those for generative AI systems.

Proposed evaluation dimensions include cross-lingual detection parity, cultural content coverage, and demographic disaggregation.

Abstract

Watermarking is becoming the default mechanism for AI content authentication, with governance policies and frameworks referencing it as infrastructure for content provenance. Yet across text, image, and audio modalities, watermark signal strength, detectability, and robustness depend on statistical properties of the content itself, properties that vary systematically across languages, cultural visual traditions, and demographic groups. We examine how this content dependence creates modality-specific pathways to bias. Reviewing the major watermarking benchmarks across modalities, we find that, with one exception, none report performance across languages, cultural content types, or population groups. To address this, we propose three concrete evaluation dimensions for pluralistic watermark benchmarking: cross-lingual detection parity, culturally diverse content coverage, and demographic disaggregation of detection metrics. We connect these to the governance frameworks currently mandating watermarking deployment and show that watermarking is held to a lower fairness standard than the generative systems it is meant to govern. Our position is that evaluation must precede deployment, and that the same bias auditing requirements applied to AI models should extend to the verification layer.