NomicLaw: Emergent Trust and Strategic Argumentation in LLMs During Collaborative Law-Making

TL;DR

This paper introduces NomicLaw, a multi-agent simulation where LLMs collaboratively engage in legal reasoning, revealing their social and strategic capabilities in open-ended legal deliberations.

Contribution

It presents a novel structured simulation framework for studying LLMs in multi-agent legal decision-making and analyzes their emergent social behaviors.

Findings

LLMs form alliances and betray trust in legal negotiations.

Agents use strategic language to influence collective decisions.

The study reveals social reasoning and persuasive skills of open-source LLMs.

Abstract

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.