Outraged AI: Large language models prioritise emotion over cost in fairness enforcement

TL;DR

This study investigates how large language models (LLMs) use emotion in fairness decisions, revealing they prioritize emotion over cost, unlike humans, with implications for developing more human-like AI moral reasoning.

Contribution

The paper provides the first causal evidence of emotion-driven moral decisions in LLMs and compares their behavior to humans, highlighting key differences and areas for improvement.

Findings

LLMs use emotion to guide punishment more strongly than humans.

Prompting emotion increases punishment in LLMs.

LLMs are less sensitive to cost than humans, enforcing norms in an all-or-none manner.

Abstract

Emotions guide human decisions, but whether large language models (LLMs) use emotion similarly remains unknown. We tested this using altruistic third-party punishment, where an observer incurs a personal cost to enforce fairness, a hallmark of human morality and often driven by negative emotion. In a large-scale comparison of 4,068 LLM agents with 1,159 adults across 796,100 decisions, LLMs used emotion to guide punishment, sometimes even more strongly than humans did: Unfairness elicited stronger negative emotion that led to more punishment; punishing unfairness produced more positive emotion than accepting; and critically, prompting self-reports of emotion causally increased punishment. However, mechanisms diverged: LLMs prioritized emotion over cost, enforcing norms in an almost all-or-none manner with reduced cost sensitivity, whereas humans balanced fairness and cost. Notably, reasoning models (o3-mini, DeepSeek-R1) were more cost-sensitive and closer to human behavior than foundation models (GPT-3.5, DeepSeek-V3), yet remained heavily emotion-driven. These findings provide the first causal evidence of emotion-guided moral decisions in LLMs and reveal deficits in cost calibration and nuanced fairness judgements, reminiscent of early-stage human responses. We propose that LLMs progress along a trajectory paralleling human development; future models should integrate emotion with context-sensitive reasoning to achieve human-like emotional intelligence.