Humans learn from catastrophic mistakes not through numerical penalties, but through qualitative suffering that reshapes who they are. Current AI safety approaches replicate none of this. Reward shaping captures magnitude, not meaning. Rule-based alignment constrains behaviour, but does not change it.
We propose Emotional Cost Functions, a framework in which agents develop Qualitative Suffering States, rich narrative representations of irreversible consequences that persist forward and actively reshape character. Unlike numerical penalties, qualitative suffering states capture the meaning of what was lost, the specific void it creates, and how it changes the agent's relationship to similar future situations. Our four-component architecture - Consequence Processor, Character State, Anticipatory Scan, and Story Update is grounded in one principle. Actions cannot be undone and agents must live with what they have caused. Anticipatory dread operates through two pathways. Experiential dread arises from the agent's own lived consequences. Pre-experiential dread is acquired without direct experience, through training or inter-agent transmission. Together they mirror how human wisdom accumulates across experience and culture.
Ten experiments across financial trading, crisis support, and content moderation show that qualitative suffering produces specific wisdom rather than generalised paralysis. Agents correctly engage with moderate opportunities at 90-100% while numerical baselines over-refuse at 90%. Architecture ablation confirms the mechanism is necessary. The full system generates ten personal grounding phrases per probe vs. zero for a vanilla LLM. Statistical validation (N=10) confirms reproducibility at 80-100% consistency.
Why We Feel: Breaking Boundaries in Emotional Reasoning with Multimodal Large Language Models
Georgia Institute of Technology(佐治亚理工学院);Carnegie Mellon University(卡内基梅隆大学);University of Washington(华盛顿大学);Shenzhen Institute of Advanced Technology(深圳先进技术研究院);Shenzhen Technology University(深圳技术大学);Alibaba Group(阿里巴巴集团);Tongji University(同济大学)
Feeling Machines: Ethics, Culture, and the Rise of Emotional AI
Fraunhofer IPK(弗劳恩霍夫生产系统与设计技术研究所);TU Berlin(柏林工业大学);Université Sorbonne Paris Nord(索邦巴黎北大学);University of Augsburg(奥格斯堡大学);EM Lyon Business School(里昂商学院);EPITA(法国巴黎计算机及自动化学院);Università di Pisa(比萨大学);University of Bamberg(班贝格大学);Universidad Central de Venezuela(委内瑞拉中央大学);RWTH Aachen University(亚琛工业大学);Harz University of Applied Sciences(哈茨应用技术大学);University of Passau(帕绍大学);SCAI, Sorbonne Université(索邦大学人工智能中心);University of Hamburg(汉堡大学);Osnabrueck University(奥斯纳布吕克大学);ICM, LIP6 UMR 7606 CNRS(法国国家科学研究中心皮埃尔和玛丽居里大学计算机科学实验室);Sorbonne Université(索邦大学)
Teaching AI to Feel: A Collaborative, Full-Body Exploration of Emotive Communication
Institute of Neurosciences of the University of Barcelona Spain;Artificial Intelligence Research Institute (IIIA-CSIC) Barcelona Spain;Massachusetts Institute of Technology Cambridge USA;ESPRONCEDA Institute of Art \& Culture Barcelona Spain;Institute of Neurosciences of the University of;Artificial Intelligence Research Institute (IIIA-CSIC);Massachusetts Institute of Technology;ESPRONCEDA Institute of Art \& Culture
Your Robot Will Feel You Now: Empathy in Robots and Embodied Agents
School of Computing Science, Simon Fraser University(计算科学学院,西蒙弗雷泽大学);School of Interactive Arts and Technology, Simon Fraser University(交互艺术与技术学院,西蒙弗雷泽大学)
EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition
LAION e.V.(LAION研究所);Technical University of Munich(慕尼黑技术大学);L3S Research Center Leibniz University of Hannover(汉诺威莱布尼茨大学L3S研究中心);TU Darmstadt(达姆施塔特技术大学);DFKI(德意志联邦防务研究所);TIB–Leibniz Information Centre for Science and Technology L3S Research Center Leibniz University of Hannover(莱布尼茨科学与技术信息中心L3S研究中心汉诺威莱布尼茨大学)