NomicLaw: Emergent Trust and Strategic Argumentation in LLMs During Collaborative Law-Making
This work addresses the need for insights into AI systems capable of autonomous negotiation and legislation in legal settings, though it is incremental in extending existing multi-agent simulations to a specific domain.
The paper tackled the problem of limited empirical understanding of LLM behavior in open-ended, multi-agent legal settings by introducing NomicLaw, a simulation where LLMs engage in collaborative law-making, and found that agents spontaneously form alliances, betray trust, and adapt rhetoric to influence decisions, highlighting latent social reasoning and persuasive capabilities in ten open-source LLMs.
Recent advancements in large language models (LLMs) have extended their capabilities from basic text processing to complex reasoning tasks, including legal interpretation, argumentation, and strategic interaction. However, empirical understanding of LLM behavior in open-ended, multi-agent settings especially those involving deliberation over legal and ethical dilemmas remains limited. We introduce NomicLaw, a structured multi-agent simulation where LLMs engage in collaborative law-making, responding to complex legal vignettes by proposing rules, justifying them, and voting on peer proposals. We quantitatively measure trust and reciprocity via voting patterns and qualitatively assess how agents use strategic language to justify proposals and influence outcomes. Experiments involving homogeneous and heterogeneous LLM groups demonstrate how agents spontaneously form alliances, betray trust, and adapt their rhetoric to shape collective decisions. Our results highlight the latent social reasoning and persuasive capabilities of ten open-source LLMs and provide insights into the design of future AI systems capable of autonomous negotiation, coordination and drafting legislation in legal settings.