Advancing AI Negotiations: New Theory and Evidence from a Large-Scale Autonomous Negotiations Competition

Despite the rapid proliferation of artificial intelligence (AI) negotiation agents, there has been limited integration of computer science research and established negotiation theory to develop new theories of AI negotiation. To bridge this gap, we conducted an International AI Negotiations Competition in which participants with a diverse range of experience in negotiation, artificial intelligence (AI), and computer science (CS) iteratively designed and refined prompts for large language model (LLM) negotiation agents. We then facilitated ~120,000 negotiations between these agents across multiple scenarios with diverse characteristics and objectives. Our findings revealed that fundamental principles from established human-human negotiation theory remain crucial in AI-AI negotiations. Specifically, agents exhibiting high “warmth” reached deals more frequently, created more value in integrative settings, claimed more value in distributive settings, and fostered higher subjective value. In addition, highly “dominant” agents performed better at claiming value in both distributive and integrative settings. These results align with classic negotiation theory emphasizing the importance of relationship-building, assertiveness, and preparation. However, our analysis also revealed unique dynamics in AI-AI negotiations not fully explained by negotiation theory, particularly regarding the effectiveness of AI-specific technical strategies like chain-of-thought reasoning and prompt injection. In fact, the agent that demonstrated the best combined performance across our key metrics—value creation, value claiming, and counterpart subjective value—implemented a sophisticated approach that blended traditional negotiation preparation frameworks with AI-specific technical methods. Together, these results suggest the importance of establishing a new theory of AI negotiations which integrates established negotiation theory with AI-specific negotiating strategies to optimize agent performance. Our research suggests this new theory must account for the unique characteristics of autonomous agents and establish the conditions under which traditional negotiation theory applies in automated settings.