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

• URL: http://arxiv.org/abs/2510.17880v1
• Date: Fri, 17 Oct 2025 08:41:36 GMT
• Title: Outraged AI: Large language models prioritise emotion over cost in fairness enforcement
• Authors: Hao Liu, Yiqing Dai, Haotian Tan, Yu Lei, Yujia Zhou, Zhen Wu,
• Abstract summary: We show that large language models (LLMs) use emotion to guide punishment.<n> Unfairness elicited stronger negative emotion that led to more punishment.<n>We propose that future models should integrate emotion with context-sensitive reasoning to achieve human-like emotional intelligence.
• Score: 13.51400164704227
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Emotionally Charged, Logically Blurred: AI-driven Emotional Framing Impairs Human Fallacy Detection [25.196971926947906]
  We present the first computational study of how emotional framing interacts with fallacies and convincingness.<n>We use large language models (LLMs) to systematically change emotional appeals in fallacious arguments.<n>Our work has implications for AI-driven emotional manipulation in the context of fallacious argumentation.
  arXiv  Detail & Related papers  (2025-10-09T14:57:37Z)
• Large Language Models are Highly Aligned with Human Ratings of Emotional Stimuli [0.62914438169038]
  Emotions exert an immense influence over human behavior and cognition in both commonplace and high-stress tasks.<n>Discussions should be informed by an understanding of how large language models evaluate emotionally loaded stimuli or situations.<n>A model's alignment with human behavior in these cases can inform the effectiveness of LLMs for certain roles or interactions.
  arXiv  Detail & Related papers  (2025-08-19T19:22:00Z)
• AI shares emotion with humans across languages and cultures [12.530921452568291]
  We assess human-AI emotional alignment across linguistic-cultural groups and model-families.<n>Our analyses reveal that LLM-derived emotion spaces are structurally congruent with human perception.<n>We show that model expressions can be stably and naturally modulated across distinct emotion categories.
  arXiv  Detail & Related papers  (2025-06-11T14:42:30Z)
• UDDETTS: Unifying Discrete and Dimensional Emotions for Controllable Emotional Text-to-Speech [61.989360995528905]
  We propose UDDETTS, a universal framework unifying discrete and dimensional emotions for controllable emotional TTS.<n>This model introduces the interpretable Arousal-Dominance-Valence (ADV) space for dimensional emotion description and supports emotion control driven by either discrete emotion labels or nonlinearly quantified ADV values.<n>Experiments show that UDDETTS achieves linear emotion control along three interpretable dimensions, and exhibits superior end-to-end emotional speech synthesis capabilities.
  arXiv  Detail & Related papers  (2025-05-15T12:57:19Z)
• Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Models [75.85319609088354]
  Sentient Agent as a Judge (SAGE) is an evaluation framework for large language models.<n>SAGE instantiates a Sentient Agent that simulates human-like emotional changes and inner thoughts during interaction.<n>SAGE provides a principled, scalable and interpretable tool for tracking progress toward genuinely empathetic and socially adept language agents.
  arXiv  Detail & Related papers  (2025-05-01T19:06:10Z)
• Human-like Affective Cognition in Foundation Models [28.631313772625578]
  We introduce an evaluation framework for testing affective cognition in foundation models. We generate 1,280 diverse scenarios exploring relationships between appraisals, emotions, expressions, and outcomes. Our results show foundation models tend to agree with human intuitions, matching or exceeding interparticipant agreement.
  arXiv  Detail & Related papers  (2024-09-18T06:42:13Z)
• GPT-4 Emulates Average-Human Emotional Cognition from a Third-Person Perspective [1.642094639107215]
  We first look at carefully crafted emotion-evoking stimuli, originally designed to find patterns of brain neural activity. We show that GPT-4 is especially accurate in reasoning about such stimuli. We find that GPT-4's interpretations align more closely with human judgments about the emotions of others than with self-assessments.
  arXiv  Detail & Related papers  (2024-08-11T01:22:09Z)
• The Good, The Bad, and Why: Unveiling Emotions in Generative AI [73.94035652867618]
  We show that EmotionPrompt can boost the performance of AI models while EmotionAttack can hinder it. EmotionDecode reveals that AI models can comprehend emotional stimuli akin to the mechanism of dopamine in the human brain.
  arXiv  Detail & Related papers  (2023-12-18T11:19:45Z)
• Language Models (Mostly) Do Not Consider Emotion Triggers When Predicting Emotion [87.18073195745914]
  We investigate how well human-annotated emotion triggers correlate with features deemed salient in their prediction of emotions. Using EmoTrigger, we evaluate the ability of large language models to identify emotion triggers. Our analysis reveals that emotion triggers are largely not considered salient features for emotion prediction models, instead there is intricate interplay between various features and the task of emotion detection.
  arXiv  Detail & Related papers  (2023-11-16T06:20:13Z)
• Socratis: Are large multimodal models emotionally aware? [63.912414283486555]
  Existing emotion prediction benchmarks do not consider the diversity of emotions that an image and text can elicit in humans due to various reasons. We propose Socratis, a societal reactions benchmark, where each image-caption (IC) pair is annotated with multiple emotions and the reasons for feeling them. We benchmark the capability of state-of-the-art multimodal large language models to generate the reasons for feeling an emotion given an IC pair.
  arXiv  Detail & Related papers  (2023-08-31T13:59:35Z)
• When to Make Exceptions: Exploring Language Models as Accounts of Human Moral Judgment [96.77970239683475]
  AI systems need to be able to understand, interpret and predict human moral judgments and decisions. A central challenge for AI safety is capturing the flexibility of the human moral mind. We present a novel challenge set consisting of rule-breaking question answering.
  arXiv  Detail & Related papers  (2022-10-04T09:04:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.