# What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices

arXiv: https://arxiv.org/abs/2504.03803 · HTML: https://mlsanity.com/paper/2504.03803/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 4 Apr 2025.

## Builds on (most cited first, 20 of 20)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs](https://mlsanity.com/paper/2308.13387/) (2023)
- [LLM4SR: A Survey on Large Language Models for Scientific Research](https://mlsanity.com/paper/2501.04306/) (2025)
- [The Political Preferences of LLMs](https://mlsanity.com/paper/2402.01789/) (2024)
- [Large Language Models Reflect the Ideology of their Creators](https://mlsanity.com/paper/2410.18417/) (2024)
- [LLM Censorship: A Machine Learning Challenge or a Computer Security Problem?](https://mlsanity.com/paper/2307.10719/) (2023)
- [Assessing Political Bias in Large Language Models](https://mlsanity.com/paper/2405.13041/) (2024)
- [Securing Large Language Models: Addressing Bias, Misinformation, and Prompt Attacks](https://mlsanity.com/paper/2409.08087/) (2024)
- [IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance](https://mlsanity.com/paper/2502.08395/) (2025)
- [The Model Openness Framework: Promoting Completeness and Openness for Reproducibility, Transparency, and Usability in Artificial Intelligence](https://mlsanity.com/paper/2403.13784/) (2024)
- [A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy](https://mlsanity.com/paper/2501.09431/) (2025)
- [Privacy in Fine-tuning Large Language Models: Attacks, Defenses, and Future Directions](https://mlsanity.com/paper/2412.16504/) (2024)
- [Cannot or Should Not? Automatic Analysis of Refusal Composition in IFT/RLHF Datasets and Refusal Behavior of Black-Box LLMs](https://mlsanity.com/paper/2412.16974/) (2024)
- [Pantheon 1.0, a manually verified dataset of globally famous biographies](https://mlsanity.com/paper/1502.07310/) (2015)
- [Comparative Global AI Regulation: Policy Perspectives from the EU, China, and the US](https://mlsanity.com/paper/2410.21279/) (2024)
- [Large Language Model Soft Ideologization via AI-Self-Consciousness](https://mlsanity.com/paper/2309.16167/) (2023)
- [Red AI? Inconsistent Responses from GPT3.5 Models on Political Issues in the US and China](https://mlsanity.com/paper/2312.09917/) (2023)
- [Assessing Large Language Models for Online Extremism Research: Identification, Explanation, and New Knowledge](https://mlsanity.com/paper/2408.16749/) (2024)

## Built on by (most cited first, 11 of 11)

- [Information Suppression in Large Language Models: Auditing, Quantifying, and Characterizing Censorship in DeepSeek](https://mlsanity.com/paper/2506.12349/) (2025)
- [Generative Exaggeration in LLM Social Agents: Consistency, Bias, and Toxicity](https://mlsanity.com/paper/2507.00657/) (2025)
- [Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics](https://mlsanity.com/paper/2512.16602/) (2025)
- [CoPE: A Small Language Model for Steerable and Scalable Content Labeling](https://mlsanity.com/paper/2512.18027/) (2025)
- [Behind the Refusal: Determining Guardrail Activation via Behavioral Monitoring](https://mlsanity.com/paper/2607.02121/) (2026)
- [Fair Representation in Parliamentary Summaries: Measuring and Mitigating Inclusion Bias](https://mlsanity.com/paper/2507.14221/) (2025)
- [Do Stop Me Now: Detecting Boilerplate Responses with a Single Iteration](https://mlsanity.com/paper/2510.22679/) (2025)
- [Exploring Learners' Expectations and Engagement When Collaborating with Constructively Controversial Peer Agents](https://mlsanity.com/paper/2601.13479/) (2026)
- [AuAu: A Benchmark for Auditing Authoritarian Alignment in Large Language Models](https://mlsanity.com/paper/2606.16127/) (2026)
- [Defeat Devices in AI Systems](https://mlsanity.com/paper/2606.28863/) (2026)
- [Directional Hallucinations: Ideological Drift in News-Grounded LLM Question Answering](https://mlsanity.com/paper/2607.20487/) (2026)

## Abstract (arXiv)

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.