# Seeing Like an AI: How LLMs Apply (and Misapply) Wikipedia Neutrality Norms

arXiv: https://arxiv.org/abs/2407.04183 · HTML: https://mlsanity.com/paper/2407.04183/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 92 references that are in the map.
- First posted on arXiv 4 Jul 2024.

## Builds on (most cited first, 12 of 12)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [Debating with More Persuasive LLMs Leads to More Truthful Answers](https://mlsanity.com/paper/2402.06782/) (2024)
- [Automatically Neutralizing Subjective Bias in Text](https://mlsanity.com/paper/1911.09709/) (2019)
- [Adapting Large Language Models for Content Moderation: Pitfalls in Data Engineering and Supervised Fine-tuning](https://mlsanity.com/paper/2310.03400/) (2023)
- [Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines](https://mlsanity.com/paper/2205.11558/) (2022)
- [Toxicity Detection is NOT all you Need: Measuring the Gaps to Supporting Volunteer Content Moderators](https://mlsanity.com/paper/2311.07879/) (2023)

## Built on by (most cited first, 8 of 8)

- [Plurals: A System for Guiding LLMs Via Simulated Social Ensembles](https://mlsanity.com/paper/2409.17213/) (2024)
- [MoMoE: Mixture of Moderation Experts Framework for AI-Assisted Online Governance](https://mlsanity.com/paper/2505.14483/) (2025)
- [WETBench: A Benchmark for Detecting Task-Specific Machine-Generated Text on Wikipedia](https://mlsanity.com/paper/2507.03373/) (2025)
- [Deep Value Benchmark: Measuring Whether Models Generalize Deep Values or Shallow Preferences](https://mlsanity.com/paper/2511.02109/) (2025)
- [Policy Maps: Tools for Guiding the Unbounded Space of LLM Behaviors](https://mlsanity.com/paper/2409.18203/) (2024)
- [Wikipedia in the Era of LLMs: Evolution and Risks](https://mlsanity.com/paper/2503.02879/) (2025)
- [Multilingual Reference Need Assessment System for Wikipedia](https://mlsanity.com/paper/2603.17146/) (2026)
- [From Web(logs) to Web(AI): Questions, Platforms, and Methods across Twenty Editions of ICWSM](https://mlsanity.com/paper/2610.00205/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are trained on broad corpora and then used in communities with specialized norms. Is providing LLMs with community rules enough for models to follow these norms? We evaluate LLMs' capacity to detect (Task 1) and correct (Task 2) biased Wikipedia edits according to Wikipedia's Neutral Point of View (NPOV) policy. LLMs struggled with bias detection, achieving only 64% accuracy on a balanced dataset. Models exhibited contrasting biases (some under- and others over-predicted bias), suggesting distinct priors about neutrality. LLMs performed better at generation, removing 79% of words removed by Wikipedia editors. However, LLMs made additional changes beyond Wikipedia editors' simpler neutralizations, resulting in high-recall but low-precision editing. Interestingly, crowdworkers rated AI rewrites as more neutral (70%) and fluent (61%) than Wikipedia-editor rewrites. Qualitative analysis found LLMs sometimes applied NPOV more comprehensively than Wikipedia editors but often made extraneous non-NPOV-related changes (such as grammar). LLMs may apply rules in ways that resonate with the public but diverge from community experts. While potentially effective for generation, LLMs may reduce editor agency and increase moderation workload (e.g., verifying additions). Even when rules are easy to articulate, having LLMs apply them like community members may still be difficult.