Large language models replicate and predict human cooperation across experiments in game theory
arXiv:2511.04500v3 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly deployed as decision-making agents in high-stakes domains and as imitators of human behavior in the social and behavioral sciences. Yet how closely LLMs mirror human decision-making remains poorly understood. This gap is critical: misalignment could produce harmful outcomes in practice, while failure to replicate human behavior renders LLMs ineffective as social simulators. Here, we address this gap by replicating large-scale game-theoretic experiments and by introducing a systematic prompti
The rapid advancement and deployment of LLMs make their interaction with human decision-making and social systems an urgent research area.
Understanding how LLMs replicate human cooperation is critical for their safe deployment in high-stakes domains and for their utility as social simulators.
The research provides a systematic method to evaluate and potentially align LLM behavior with human decision-making in complex social interactions.
- · AI developers
- · Social scientists
- · Organizations deploying LLMs
- · Developers ignoring ethical AI
- · Purely analytical economic modeling
Increased understanding of LLM capabilities and limitations in mimicking human social behavior.
Improved design and deployment of LLMs for tasks requiring nuanced social intelligence or human interaction.
Potential for LLMs to serve as foundational models for designing new economic systems or social policies based on predicted human responses.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL