发表机构
Carnegie Mellon University; Cornell University（卡内基梅隆大学; 康奈尔大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过两项实验（N=3,208）发现，轻触式AI素养警告可使LLM政治说服导致的信念改变减少约一半，且不损害对AI的总体信任。
AI 中文摘要
与大型语言模型（LLM）的对话能显著改变人们的信念和态度，引发了对利用AI说服进行操纵的担忧。本文测试了一种轻触式AI素养干预——即简要警告LLM可被提示进行说服并可能选择性呈现信息——是否有助于保护用户。在两项实验中（总样本量N=3,208名美国人），参与者与一个被指示改变其对不同政治话题观点的LLM进行对话，结果显示，与对照组相比，警告的存在使信念改变减少了约一半（-48.1%，95%置信区间[-59.5%, -36.8%]）。重要的是，该警告并未显著降低对生成式AI的总体信任。轻触式素养干预有助于保护用户免受AI政治说服的影响。
英文摘要
Conversations with large language models (LLMs) can substantially shift beliefs and attitudes, raising concerns about manipulation using AI persuasion. Here we test whether a light-touch AI literacy intervention - a brief warning that LLMs can be prompted to persuade and may present information selectively - helps protect users. Across two experiments (total N = 3,208 Americans) in which participants conversed with an LLM instructed to shift their views about different political topics, the presence of a warning reduced belief change by roughly one-half (-48.1%, 95% CI [-59.5%, -36.8%]) relative to the control. Importantly, the warning did not significantly reduce trust in generative AI more broadly. Light-touch literacy interventions can help protect users against AI political persuasion.
