The Art of Midwifery in LLMs: Optimizing Role Personas for Large Language Models as Moral Assistants

2026-03-20

AI Summary

This study challenges the prevailing view of large language models (LLMs) as standalone “moral agents” and instead reconceptualizes them as “moral assistants” that foster human moral reflection. Drawing on the Socratic method of maieutics, the authors design role-based personas—such as exemplars of virtue, guardian angels, and Socratic guides—that engage users through a mechanism of “constructive disagreement” across six moral scenarios. Rather than substituting human judgment, these personas offer pluralistic perspectives to stimulate autonomous ethical reasoning. Experimental results demonstrate context-dependent strengths: the virtue exemplar achieves the best overall performance, the guardian angel provides superior emotional support in bioethical crises, and the Socratic persona most effectively prompts existential reflection. This approach transcends conventional human-AI alignment paradigms by prioritizing dialogic moral engagement over prescriptive decision-making.

Technology Category

• Philosophy and Ethics of AI: Morality & Value-based AI
• Humans and AI: Learning Human Values and Preferences
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

With the development of Large Language Models (LLMs) in consulting, their role in moral decision-making has become prominent. However, existing research predominantly consider AI as an independent "moral agent" adhering to the "Human-AI Alignment" paradigm. In this study, we propose that AI should serve as a "moral assistant", facilitating users' moral growth through the "Art of Midwifery" rather than substituting human judgment. We endow LLMs with distinct persona archetypes and conducted dialogues across six moral scenarios. Findings reveal that while the virtue exemplar excelled overall, optimal performance was context-dependent: the Guardian Angel excelled in bioethical crises for emotional support, whereas the Socratic persona better elicited reflection in existential dilemmas. We introduce "Constructive Divergence", arguing that AI should offer alternative perspectives at critical moment rather than blindly accommodate users, transcending traditional alignment paradigms.

Problem

• moral assistant
• human-AI alignment
• role persona
• constructive divergence
• moral decision-making

Innovation

• Moral Assistant
• Art of Midwifery
• Constructive Divergence
• Persona Archetypes
• Human-AI Alignment