Group Selection Promotes Prosocial Prompts in Populations of LLM Agents
2026-06-22 • Computers and Society
Computers and Society
AI summaryⓘ
The authors studied how groups of language model agents can learn to cooperate better. They found that when groups are chosen based on their overall success, the agents tend to develop and share helpful, prosocial behaviors. However, if agents are selected individually for their own success, selfish behaviors take over and cooperation breaks down. Their results, supported by simulations and mathematical models, suggest that group-level competition encourages cooperation among AI agents. They also saw that a newer model, GPT-5.4, could predict and start cooperating earlier when it knew about the group selection process.
Large Language ModelsProsocialityMulti-agent systemsGroup selectionSocial dilemma gamesReplicator-mutator modelCooperationPrompt engineeringEvolutionary mechanismsAnticipatory behavior
Authors
Luis Celiktemel, Edward Eichhorn, Levin Brinkmann, Robin Schimmelpfennig, Aron Vallinder, Yaomin Jiang, Edward Hughes, Iyad Rahwan
Abstract
Current approaches to instill prosociality in large language model (LLM) agents often rely on humans specifying desired behaviors at the individual level, which does not guarantee cooperation within LLM populations. As frontier training shifts toward individual rewards for verifiable tasks, such as mathematics and coding, this outcome-based focus may further undermine cooperation in multi-agent settings. Large-scale cooperation in human populations emerged via unguided evolutionary mechanisms, not a central architect. Group selection, in which cooperative groups within a population outcompete less cooperative ones, has been argued to be essential. In this study, we explore whether group selection can promote cooperation in populations of LLM agents. We introduce a multi-agent simulation framework in which LLM agents play a repeated social dilemma game and transmit their natural-language prompts across generations under either individual- or group-level selection. Under group selection, prompts from high-performing groups are transmitted, thereby promoting prosociality and stabilizing cooperation. Under individual selection, self-interested prompts dominate, causing populations to collapse into collective defection. This gap is robust across prompt ablations, alternative game framings, and model swaps. We theoretically reproduce key results using a replicator-mutator model, whose empirical transmission kernel predicts a phase transition at a critical threshold. Preliminary findings show that, when informed about the selection mechanism, GPT-5.4 preemptively and gradually adjusts first-generation donations. This demonstrates strong anticipatory behavior that was not observed in the other tested models. These results demonstrate that prosocial prompts and cooperative behaviors evolve in LLM agent populations under group selection.