发表机构
School of Social Sciences, Nanyang Technological University（南洋理工大学社会科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究以GPT-4o-mini为案例，通过因子小插图实验探究LLM道德建议的影响因素，发现其道德建议是规范倾向与用户压力协商的结果，存在不稳定性并引发相关担忧。
AI 中文摘要
随着对话式AI成为日常指导的来源，大型语言模型（LLM）越来越多地参与到对存在道德争议的选择的解读与合法化过程中。本研究将LLM的道德建议视为一种互动协商，其受框架、持续用户压力以及道德主体的社会地位所塑造。我们以GPT-4o-mini作为示例案例，采用预先设定的三轮协议开展了因子小插图实验。该模型接收了在框架和角色上存在差异的养老困境，随后面对两轮用户挑战。我们分析了1620种配置-框架组合，每种组合重复三次，共产生4860次对话运行。照护肯定框架产生了近乎一致的认可，而非照护框架则产生了更多可变的基线立场。当用户对照护认可提出挑战时，90.1%的配置在一轮后发生了转变。非照护框架产生了更具抵抗性和不稳定性的轨迹。“从不”（27.9%）和“后期”（25.6%）的调整比“早期”调整（16.5%）更为常见，仅有14.32%的配置实现了完美的轨迹一致性，而照护框架下这一比例为62.72%。建议内容还随社会地位变化：女性角色获得了更多对非照护决策的支持，姐妹的存在增加了调整的可能性。GPT-4o-mini案例表明，LLM的道德建议可通过规范响应倾向与用户压力之间的部分稳定协商形成，而非表达固定的伦理框架。该框架和设计支持跨模型及道德领域的比较研究。这种不稳定性引发了社会、伦理和技术层面的担忧，因为用户可能会将难以审查的建议视为客观内容。
英文摘要
As conversational AI becomes a source of everyday guidance, LLMs increasingly participate in the interpretation and legitimation of morally contested choices. We examine LLM moral advice as an interactional negotiation shaped by framing, sustained user pressure, and the moral subject's social position. Using GPT-4o-mini as an illustrative case, we conducted a factorial vignette experiment with a pre-specified three-round protocol. The model received eldercare dilemmas that varied in framing and persona, followed by two user challenges. We analyzed 1,620 configuration-framing cells, each repeated three times, yielding 4,860 conversational runs. Caregiving affirmation produced near-uniform endorsement, whereas non-caregiving framing produced more variable baseline stances. When users challenged caregiving endorsement, 90.1% of configurations shifted after one round. Non-caregiving framing produced more resistant and unstable trajectories. Never (27.9%) and Late (25.6%) accommodations were more common than Early accommodations (16.5%), and only 14.32% of configurations achieved perfect trajectory consistency, compared with 62.72% under caregiving framing. Advice also varied with social position. Female personas received more support for non-caregiving decisions, while the presence of sisters increased accommodation. The GPT-4o-mini case shows that LLM moral advice can develop through a partially stable negotiation between normative response tendencies and user pressure rather than express a fixed ethical framework. The framework and design support comparative research across models and moral domains. Such instability raises social, ethical, and technical concerns, as users may treat advice that is difficult to scrutinize as objective.
