Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption
Yepeng Liu, Xuandong Zhao, Dawn Song, Gregory W. Wornell, Yuheng Bu

TL;DR
This paper argues that aligning stakeholder incentives is crucial for the practical adoption of LLM watermarking, proposing in-context watermarking as a promising approach for trusted parties.
Contribution
It introduces the concept of incentive-aligned watermarking, especially in trusted domains, and provides design principles and future directions for practical deployment.
Findings
In-context watermarking enables detection without quality loss.
Aligning incentives encourages adoption among stakeholders.
Different watermarking methods suit various application domains.
Abstract
Despite progress in watermarking algorithms for large language models (LLMs), real-world deployment remains limited. We argue that this gap stems from misaligned incentives among LLM providers, platforms, and end users, which manifest as three key barriers: competitive risk, detection-tool governance, and attribution issues. We revisit three classes of watermarking through this lens. \emph{Model watermarking} naturally aligns with LLM provider interests, yet faces new challenges in open-source ecosystems. \emph{LLM text watermarking} offers modest provider benefit when framed solely as an anti-misuse tool, but can gain traction in narrowly scoped settings such as dataset de-contamination or user-controlled provenance. \emph{In-context watermarking} (ICW) is tailored for trusted parties, such as conference organizers or educators, who embed hidden watermarking instructions into…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
