arXiv cs.CY· Kobi Hackenburg, Caroline Wagner, Luke Hewitt, Ben M. Tappin, Ed Saunders, Hannah Rose Kirk, Helen Margetts, Christopher Summerfield·· 10 小时前精选AI 评分76
arXiv 论文：AI 在文本对话中的说服力超过人类专家
AI out-persuades expert humans in text-based dialogue
AI 导读
一项预注册研究通过四项实验（共 6,923 人、18,978 段对话）发现，AI 系统在文本对话中的说服力稳定超过人类专家，对手包括线上说服锦标赛冠军、职业拉票员和世界辩论冠军，即便专家自选议题、提前调研、接受数小时实战训练并获得 1,000 英镑奖金，仍未能取胜。
推荐理由
四项预注册实验覆盖近1.9万段对话，并延伸到真实募捐场景，为AI说服力优势提供了可比较的实证依据。
正文 · 原文
Abstract:Many societal decisions are settled by contests of persuasion. Conversational AI is a powerful new entrant in these contests, but whether it can out-persuade skilled and highly incentivized humans has remained unclear. Here, in a series of four preregistered experiments (n = 18,978 conversations from 6,923 people), we pitted AI systems against a range of human persuaders, including laypeople, winners of a separately preregistered four-round online persuasion tournament, professional canvassers, and world championship debaters. We found that AI systems were reliably more persuasive than expert humans, even when expert humans chose their issues, researched in advance, underwent hours of live, structured practice, and were incentivized with £1,000 cash bonuses. In a follow-up study, AI's advantage persisted after experts received a coaching tool that let them practice against the AI that beat them, review their performance history, and see what AI would have said at key moments. We found converging evidence that AI's advantage stemmed from rapidly deploying larger quantities of information: after coaching, expert humans could tie an AI constrained to respond at human speeds and with human-length messages. In a final study, we show that AI's advantage extends to consequential real-world behavior: AI was more than 3x more effective than professional canvassers from a UK fundraising firm at raising real-money donations to Save the Children. Together, these results establish that frontier AI systems out-persuade expert humans in text-based conversation, with significant implications for political communication.
| Comments: | 16 pages main text, 113 pages Supplementary Information. Data and code: this https URL (doi:https://doi.org/10.5281/zenodo.20629034) |
| Subjects: | Computers and Society (cs.CY); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.16475 [cs.CY] |
| (or arXiv:2606.16475v2 [cs.CY] for this version) | |
| https://doi.org/10.48550/arXiv.2606.16475 arXiv-issued DOI via DataCite |
Submission history
From: Kobi Hackenburg [view email]
[v1]
Mon, 15 Jun 2026 09:40:28 UTC (143 KB)
[v2]
Fri, 6 Nov 2026 12:15:01 UTC (1,503 KB)
来源：arXiv cs.CY · arxiv.org