# On-Device Watermarking: A Socio-Technical Imperative For Authenticity In The Age of Generative AI

arXiv: https://arxiv.org/abs/2504.13205 · HTML: https://mlsanity.com/paper/2504.13205/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security.
- Builds on 9 of its 24 references that are in the map.
- First posted on arXiv 15 Apr 2025.

## Placed in

- LLM safety and security / [Watermarking generated content and models](https://mlsanity.com/safety/watermarking-generated-content-and-models/)

## Builds on (most cited first, 9 of 9)

- [Can AI-Generated Text be Reliably Detected?](https://mlsanity.com/paper/2303.11156/) (2023)
- [The Stable Signature: Rooting Watermarks in Latent Diffusion Models](https://mlsanity.com/paper/2303.15435/) (2023)
- [GPT detectors are biased against non-native English writers](https://mlsanity.com/paper/2304.02819/) (2023)
- [Watermarks in the Sand: Impossibility of Strong Watermarking for Generative Models](https://mlsanity.com/paper/2311.04378/) (2023)
- [Robustness of AI-Image Detectors: Fundamental Limits and Practical Attacks](https://mlsanity.com/paper/2310.00076/) (2023)
- [Evading Watermark based Detection of AI-Generated Content](https://mlsanity.com/paper/2305.03807/) (2023)
- [A Systematic Review on Model Watermarking for Neural Networks](https://mlsanity.com/paper/2009.12153/) (2020)
- [Certifiably Robust Image Watermark](https://mlsanity.com/paper/2407.04086/) (2024)
- [Generative Models are Self-Watermarked: Declaring Model Authentication through Re-Generation](https://mlsanity.com/paper/2402.16889/) (2024)

## Abstract (arXiv)

As generative AI models produce increasingly realistic output, both academia and industry are focusing on the ability to detect whether an output was generated by an AI model or not. Many of the research efforts and policy discourse are centered around robust watermarking of AI outputs. While plenty of progress has been made, all watermarking and AI detection techniques face severe limitations. In this position paper, we argue that we are adopting the wrong approach, and should instead focus on watermarking via cryptographic signatures trustworthy content rather than AI generated ones. For audio-visual content, in particular, all real content is grounded in the physical world and captured via hardware sensors. This presents a unique opportunity to watermark at the hardware layer, and we lay out a socio-technical framework and draw parallels with HTTPS certification and Blu-Ray verification protocols. While acknowledging implementation challenges, we contend that hardware-based authentication offers a more tractable path forward, particularly from a policy perspective. As generative models approach perceptual indistinguishability, the research community should be wary of being overly optimistic with AI watermarking, and we argue that AI watermarking research efforts are better spent in the text and LLM space, which are ultimately not traceable to a physical sensor.