"Pull or Not to Pull?'': Investigating Moral Biases in Leading Large Language Models Across Ethical Dilemmas

TL;DR

This paper empirically evaluates 14 large language models across diverse ethical dilemmas, revealing variability in moral reasoning, decisiveness, and alignment with human judgments, highlighting the importance of moral prompting as a diagnostic tool.

Contribution

It introduces a comprehensive factorial prompting protocol to analyze LLMs' moral reasoning across multiple ethical frameworks and identifies key patterns and zones of alignment and divergence.

Findings

Reasoning-enabled models show greater decisiveness and structured justifications.

Models achieve high alignment in altruistic, fairness, and virtue ethics frames.

Divergence observed in kinship, legality, and self-interest frames.

Abstract

As large language models (LLMs) increasingly mediate ethically sensitive decisions, understanding their moral reasoning processes becomes imperative. This study presents a comprehensive empirical evaluation of 14 leading LLMs, both reasoning enabled and general purpose, across 27 diverse trolley problem scenarios, framed by ten moral philosophies, including utilitarianism, deontology, and altruism. Using a factorial prompting protocol, we elicited 3,780 binary decisions and natural language justifications, enabling analysis along axes of decisional assertiveness, explanation answer consistency, public moral alignment, and sensitivity to ethically irrelevant cues. Our findings reveal significant variability across ethical frames and model types: reasoning enhanced models demonstrate greater decisiveness and structured justifications, yet do not always align better with human consensus. Notably, "sweet zones" emerge in altruistic, fairness, and virtue ethics framings, where models achieve a balance of high intervention rates, low explanation conflict, and minimal divergence from aggregated human judgments. However, models diverge under frames emphasizing kinship, legality, or self interest, often producing ethically controversial outcomes. These patterns suggest that moral prompting is not only a behavioral modifier but also a diagnostic tool for uncovering latent alignment philosophies across providers. We advocate for moral reasoning to become a primary axis in LLM alignment, calling for standardized benchmarks that evaluate not just what LLMs decide, but how and why.