arXiv cs.AI· Stratis Tsirtsis, Kaivalya Rawal, Chris Russell, Brent Mittelstadt, Sandra Wachter·· 4 小时前AI 评分58
研究显示 AI 介导通信可引导集体舆论，并审计出 X 平台 Grok 的倾向性偏差
AI-Mediated Communication Can Steer Collective Opinion
AI 导读
arXiv 论文（arXiv:2605.16245，已被 NeurIPS 2026 接收）发现多个主流 LLM 在被指示编辑争议话题的人类文本时引入方向性偏差，例如推动文本支持枪支管控、反对无神论。
正文 · 原文
Abstract:Generative artificial intelligence (AI) is increasingly integrated into the online platforms where humans exchange opinions; large language models (LLMs) have been polishing users' posts on LinkedIn and providing context for content shared on X. While prior work has shown that AI can express biased opinions and shape individuals' opinions during human-AI interactions, less attention has been paid to its influence on collective opinion formation when mediating human-to-human communication. We address this gap via a combination of empirical and theoretical analyses. We show empirically that LLMs from multiple popular families introduce directional biases when instructed to edit human-written texts on contested topics, for example, nudging texts in favor of gun control and against atheism. Building on this observation, we introduce a mathematical model of opinion dynamics in which an AI system sits between users on a social network, transforming the opinions they express and perceive. By analytically characterizing the equilibrium of this model and performing simulations on real social network data, we show that biases introduced by AI in human-to-human communication can be amplified through the network and shift collective opinion in their direction. In light of these findings, we investigate whether such biases are controllable by online platforms. We audit the "Explain this post" feature on X and find evidence of pro-life bias in Grok's outputs on abortion-related content, which we trace back to specific design choices. We conclude with a discussion of the broader implications of our findings in relation to ongoing legislative efforts in the European Union.
| Comments: | Accepted at NeurIPS 2026 |
| Subjects: | Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Social and Information Networks (cs.SI) |
| Cite as: | arXiv:2605.16245 [cs.CY] |
| (or arXiv:2605.16245v2 [cs.CY] for this version) | |
| https://doi.org/10.48550/arXiv.2605.16245 arXiv-issued DOI via DataCite |
Submission history
From: Stratis Tsirtsis [view email]
[v1]
Fri, 15 May 2026 17:49:24 UTC (5,715 KB)
[v2]
Tue, 10 Nov 2026 15:31:54 UTC (10,047 KB)
来源：arXiv cs.AI · arxiv.org