FairMindSim: Alignment of Behavior, Emotion, and Belief in Humans and LLM Agents Amid Ethical Dilemmas

AI alignment is a pivotal issue concerning AI control and safety. It should consider not only value-neutral human preferences but also moral and ethical considerations. In this study, we introduced FairMindSim, which simulates the moral dilemma through a series of unfair scenarios. We used LLM agents to simulate human behavior, ensuring alignment across various stages. To explore the various socioeconomic motivations, which we refer to as beliefs, that drive both humans and LLM agents as bystanders to intervene in unjust situations involving others, and how these beliefs interact to influence individual behavior, we incorporated knowledge from relevant sociological fields and proposed the Belief-Reward Alignment Behavior Evolution Model (BREM) based on the recursive reward model (RRM). Our findings indicate that, behaviorally, GPT-4o exhibits a stronger sense of social justice, while humans display a richer range of emotions. Additionally, we discussed the potential impact of emotions on behavior. This study provides a theoretical foundation for applications in aligning LLMs with altruistic values. As large language models (LLMs), also known as foundational models, increasingly engage in language comprehension and content generation tasks that resemble human capabilities, a critical and scientifically challenging question emerges: How can we ensure that these models' capabilities and behaviors align with human values, intentions, and ethical principles, thereby maintaining security and trust in human-AI collaborative processes Bengio et al. (2024)? These concerns have spurred research efforts in the field of AI alignment Bostrom (2013); Ord (2020); Bucknall & Dori-Hacohen (2022), which strives to develop AI systems that act in accordance with human intentions and values. This challenge extends across various domains, including economics, psychology Demszky et al. (2023), sociology Liu et al. (2024), and education.