arXiv cs.CL· David Gottlieb, Jared Moore·· 10 小时前AI 评分22
如何构建道德智能体:以情感主义与理性主义为视角审视 AI
How to make a moral agent: sentimentalism and rationalism as lenses on AI
AI 导读
针对 AI 系统能否成为道德主体这一尚无定论的问题,研究者提出一套描述性道德主体研究纲领:先从哲学与道德心理学文献中提取人类道德能力候选清单,再将其细化为可在 AI 系统中检验的能力,最后给出针对 LLM 的具体实证测试建议。该路径可用于评估 AI 行为是否对齐、辅助模型部署的政策判断,并反哺模型设计。
正文 · 原文
Abstract:Can AI systems be moral agents? That is, can they undertake actions that can be correctly assessed as morally right or wrong? There is no settled normative criterion of moral agency that applies neatly outside paradigm human cases. Because of the difficulties with a normative account, we instead propose a research program to explore descriptive moral agency. Most basically, we ask, what are the capabilities we humans possess that underlie our moral agency? To do so, we draw on historical scholarship in philosophy and moral psychology to isolate candidate capabilities. Then we suggest refining candidate capabilities so that they are testable in the context of AI systems. Finally, we offer specific empirical recommendations to test AI systems, especially large language models (LLMs), for those refined capabilities. Our approach offers several benefits. Firstly, it may demonstrate whether the behaviors of AI systems are aligned, or in a broad sense desirable. Secondly, the debate may inform policy questions (e.g., what models should be deployed in which contexts?). Thirdly, our approach may complement engineering and design efforts: when capabilities can be precisely characterized, they can inform the development of models. Fourthly, applying our moral reasoning concepts in the new context of AI systems should improve our understanding of those concepts.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.18714 [cs.AI] |
| (or arXiv:2611.18714v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.18714 arXiv-issued DOI via DataCite |
Submission history
From: Jared Moore [view email]
[v1]
Sat, 7 Nov 2026 22:47:19 UTC (42 KB)
[v2]
Thu, 19 Nov 2026 16:20:29 UTC (42 KB)
来源:arXiv cs.CL · arxiv.org