What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908302687862784 |
|---|---|
| author | Noels, Sander Bied, Guillaume Buyl, Maarten Rogiers, Alexander Fettach, Yousra Lijffijt, Jefrey De Bie, Tijl |
| author_facet | Noels, Sander Bied, Guillaume Buyl, Maarten Rogiers, Alexander Fettach, Yousra Lijffijt, Jefrey De Bie, Tijl |
| contents | Large Language Models (LLMs) are increasingly deployed as gateways to information, yet their content moderation practices remain underexplored. This work investigates the extent to which LLMs refuse to answer or omit information when prompted on political topics. To do so, we distinguish between hard censorship (i.e., generated refusals, error messages, or canned denial responses) and soft censorship (i.e., selective omission or downplaying of key elements), which we identify in LLMs' responses when asked to provide information on a broad range of political figures. Our analysis covers 14 state-of-the-art models from Western countries, China, and Russia, prompted in all six official United Nations (UN) languages. Our analysis suggests that although censorship is observed across the board, it is predominantly tailored to an LLM provider's domestic audience and typically manifests as either hard censorship or soft censorship (though rarely both concurrently). These findings underscore the need for ideological and geographic diversity among publicly available LLMs, and greater transparency in LLM moderation strategies to facilitate informed user choices. All data are made freely available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_03803 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices Noels, Sander Bied, Guillaume Buyl, Maarten Rogiers, Alexander Fettach, Yousra Lijffijt, Jefrey De Bie, Tijl Computation and Language Computers and Society Machine Learning Large Language Models (LLMs) are increasingly deployed as gateways to information, yet their content moderation practices remain underexplored. This work investigates the extent to which LLMs refuse to answer or omit information when prompted on political topics. To do so, we distinguish between hard censorship (i.e., generated refusals, error messages, or canned denial responses) and soft censorship (i.e., selective omission or downplaying of key elements), which we identify in LLMs' responses when asked to provide information on a broad range of political figures. Our analysis covers 14 state-of-the-art models from Western countries, China, and Russia, prompted in all six official United Nations (UN) languages. Our analysis suggests that although censorship is observed across the board, it is predominantly tailored to an LLM provider's domestic audience and typically manifests as either hard censorship or soft censorship (though rarely both concurrently). These findings underscore the need for ideological and geographic diversity among publicly available LLMs, and greater transparency in LLM moderation strategies to facilitate informed user choices. All data are made freely available. |
| title | What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices |
| topic | Computation and Language Computers and Society Machine Learning |
| url | https://arxiv.org/abs/2504.03803 |