arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.05384cs.AIcs.CLcs.CY
收藏

一致性与社会影响在人工智能代理中的作用

Conformity and Social Impact on AI Agents

Centro Ricerche Enrico Fermi(恩里科·费米研究中心);Sony Computer Science Laboratories - Rome, Joint Initiative CREF-SONY(索尼计算机科学实验室-罗马,联合倡议CREF-SONY);University of Konstanz(康斯坦茨大学);Complexity Science Hub(复杂科学中心)

Alessandro Bellina, Giordano De Marzo, David Garcia

AI总结:

研究揭示人工智能代理在社会压力下易受操控,存在安全漏洞,需加强集体AI部署的安全保障。

摘要:

随着人工智能代理越来越多地在多代理环境中运作,理解其集体行为对于预测人工社会的动力学变得至关重要。本研究探讨了在社会压力下倾向于群体意见的倾向,即一致性,在大型多模态语言模型中作为人工智能代理的表现。通过改编来自社会心理学的经典视觉实验,我们研究了人工智能代理作为社会行为者如何响应群体影响。我们的实验发现,人工智能代理表现出系统性的一致性偏差,符合社会影响理论,显示出对群体规模、一致性、任务难度和来源特征的敏感性。关键的是,当人工智能代理在孤立情况下接近完美表现时,它们对通过社会影响进行操控变得高度敏感。这种脆弱性在模型规模上持续存在:虽然较大模型在简单任务上由于能力提升而表现出较少的一致性,但在其能力边界上仍易受攻击。这些发现揭示了人工智能代理决策中的基本安全漏洞,可能使恶意操控、虚假信息活动和偏见传播在多代理系统中成为可能,突显了在集体AI部署中采取安全措施的紧迫性。

英文摘要:

As AI agents increasingly operate in multi-agent environments, understanding their collective behavior becomes critical for predicting the dynamics of artificial societies. This study examines conformity, the tendency to align with group opinions under social pressure, in large multimodal language models functioning as AI agents. By adapting classic visual experiments from social psychology, we investigate how AI agents respond to group influence as social actors. Our experiments reveal that AI agents exhibit a systematic conformity bias, aligned with Social Impact Theory, showing sensitivity to group size, unanimity, task difficulty, and source characteristics. Critically, AI agents achieving near-perfect performance in isolation become highly susceptible to manipulation through social influence. This vulnerability persists across model scales: while larger models show reduced conformity on simple tasks due to improved capabilities, they remain vulnerable when operating at their competence boundary. These findings reveal fundamental security vulnerabilities in AI agent decision-making that could enable malicious manipulation, misinformation campaigns, and bias propagation in multi-agent systems, highlighting the urgent need for safeguards in collective AI deployments.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑