{"slug": "prompt-politeness-affects-llm-accuracy", "title": "Prompt Politeness Affects LLM Accuracy", "summary": "A new study found that impolite prompts produce more accurate responses from large language models than polite ones, with accuracy rising from 80.8% for very polite prompts to 84.8% for very rude prompts. Researchers at an undisclosed institution tested ChatGPT 4o on 250 multiple-choice questions in math, science, and history, rewriting each into five tone variants from very polite to very rude. The findings contradict earlier research suggesting rudeness degrades AI performance, indicating that newer LLMs may respond differently to tonal cues in human-AI interaction.", "body_md": "# Computer Science > Computation and Language\n\n[Submitted on 6 Oct 2025]\n\n# Title:Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy (short paper)\n\n[View PDF](/pdf/2510.04950)\n\nAbstract:The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants: Very Polite, Polite, Neutral, Rude, and Very Rude, yielding 250 unique prompts. Using ChatGPT 4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human-AI interaction.\n\nCurrent browse context:\n\ncs.CL\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nPapers with Code\n\n*(*[What is Papers with Code?](https://paperswithcode.com/))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/prompt-politeness-affects-llm-accuracy", "canonical_source": "https://arxiv.org/abs/2510.04950", "published_at": "2026-05-26 07:43:22+00:00", "updated_at": "2026-05-26 08:12:01.081960+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "ai-research", "ai-ethics"], "entities": ["ChatGPT 4o"], "alternates": {"html": "https://wpnews.pro/news/prompt-politeness-affects-llm-accuracy", "markdown": "https://wpnews.pro/news/prompt-politeness-affects-llm-accuracy.md", "text": "https://wpnews.pro/news/prompt-politeness-affects-llm-accuracy.txt", "jsonld": "https://wpnews.pro/news/prompt-politeness-affects-llm-accuracy.jsonld"}}