Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest

Language Model (LM)-based agents remain largely untested in mixed-motive settings where agents must leverage short-term cooperation for long-term competitive goals (e.g., multi-party politics). We introduce Cooperate to Compete (C2C), a multi-agent environment where players can engage in private negotiations while competing to be the first to achieve their secret objective. Players have asymmetric objectives and negotiations are non-binding, allowing alliances to form and break as players' short-term interests align and diverge. We run AI-only games and conduct a user study pitting human players against AI opponents. We identify significant differences between human and AI negotiation behaviors, finding that humans favor lower-complexity deals and the top performing humans follows-through on agreements significantly less than the top performing AI model. We also find that humans are more aggressive negotiators, accepting deals without a counteroffer only 56.3\% of the time compared to 67.6\% for average LM-based agents. Through targeted interventions inspired by these differences, we modify agents' negotiation behavior and improve win rates from 22.2\% to 32.7\%. We run over 1,100 games with over 16,000 private conversations totaling 15.2 million tokens and over 150,000 player actions. Our results establish C2C as a rigorous testbed for studying and building LM-based agents that can navigate the sophisticated coordination required for real-world deployments.