Watermarking Large Language Models in Europe: Interpreting the AI Act in Light of Technology

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Souverain, Thomas
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918187405148160
author Souverain, Thomas
author_facet Souverain, Thomas
contents To foster trustworthy Artificial Intelligence (AI) within the European Union, the AI Act requires providers to mark and detect the outputs of their general-purpose models. The Article 50 and Recital 133 call for marking methods that are ''sufficiently reliable, interoperable, effective and robust''. Yet, the rapidly evolving and heterogeneous landscape of watermarks for Large Language Models (LLMs) makes it difficult to determine how these four standards can be translated into concrete and measurable evaluations. Our paper addresses this challenge, anchoring the normativity of European requirements in the multiplicity of watermarking techniques. Introducing clear and distinct concepts on LLM watermarking, our contribution is threefold. (1) Watermarking Categorisation: We propose an accessible taxonomy of watermarking methods according to the stage of the LLM lifecycle at which they are applied - before, during, or after training, and during next-token distribution or sampling. (2) Watermarking Evaluation: We interpret the EU AI Act's requirements by mapping each criterion with state-of-the-art evaluations on robustness and detectability of the watermark, and of quality of the LLM. Since interoperability remains largely untheorised in LLM watermarking research, we propose three normative dimensions to frame its assessment. (3) Watermarking Comparison: We compare current watermarking methods for LLMs against the operationalised European criteria and show that no approach yet satisfies all four standards. Encouraged by emerging empirical tests, we recommend further research into watermarking directly embedded within the low-level architecture of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03641
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Watermarking Large Language Models in Europe: Interpreting the AI Act in Light of Technology
Souverain, Thomas
Cryptography and Security
Artificial Intelligence
Computation and Language
Computers and Society
68T01, 68727, 68T30, 68T35, 68T37, 68T50
To foster trustworthy Artificial Intelligence (AI) within the European Union, the AI Act requires providers to mark and detect the outputs of their general-purpose models. The Article 50 and Recital 133 call for marking methods that are ''sufficiently reliable, interoperable, effective and robust''. Yet, the rapidly evolving and heterogeneous landscape of watermarks for Large Language Models (LLMs) makes it difficult to determine how these four standards can be translated into concrete and measurable evaluations. Our paper addresses this challenge, anchoring the normativity of European requirements in the multiplicity of watermarking techniques. Introducing clear and distinct concepts on LLM watermarking, our contribution is threefold. (1) Watermarking Categorisation: We propose an accessible taxonomy of watermarking methods according to the stage of the LLM lifecycle at which they are applied - before, during, or after training, and during next-token distribution or sampling. (2) Watermarking Evaluation: We interpret the EU AI Act's requirements by mapping each criterion with state-of-the-art evaluations on robustness and detectability of the watermark, and of quality of the LLM. Since interoperability remains largely untheorised in LLM watermarking research, we propose three normative dimensions to frame its assessment. (3) Watermarking Comparison: We compare current watermarking methods for LLMs against the operationalised European criteria and show that no approach yet satisfies all four standards. Encouraged by emerging empirical tests, we recommend further research into watermarking directly embedded within the low-level architecture of LLMs.
title Watermarking Large Language Models in Europe: Interpreting the AI Act in Light of Technology
topic Cryptography and Security
Artificial Intelligence
Computation and Language
Computers and Society
68T01, 68727, 68T30, 68T35, 68T37, 68T50
url https://arxiv.org/abs/2511.03641