This research pioneers a new approach to studying persuasive communication strategies for promoting sustainable food choices without traditional human participant studies.

• Creates multi-round dialogues between AI agents to simulate real-world meat reduction conversations
• Measures outcomes using established psychological measures
• Reduces the need for costly human participant research while exploring effective communication strategies
• Addresses the tension between health/environmental benefits and strong social norms around meat consumption

This work offers significant implications for public health initiatives by providing scalable methods to test and refine communication approaches that could influence dietary behaviors affecting both human and planetary health.

Simulating Persuasive Dialogues on Meat Reduction with Generative Agents