Nature review maps four LLM roles in social science research

A Nature Human Behaviour review published on October 26, 2026 organizes the use of large language models in computational social science into four research roles: "subjects that simulate human behaviour, assistants that automate analysis, collaborators that generate hypotheses and evaluators that audit and improve research quality."

The lead authors, Fengli Xu, Chenyang Shao, Lin Chen, James Evans and colleagues at Tsinghua University, HKUST, Microsoft Research Asia and the University of Chicago, argue LLMs "accelerate discovery while creating new methodological paradigms" across sociology, political science, psychology, anthropology and economics. When prompted with demographic or ideological profiles, the authors report, models "reproduce survey distributions, economic games and polarization dynamics," with documented applications spanning simulations of millions of agents, polarization forecasting and market dynamics.

The review does not soften the limits. LLMs "struggle with causal inference and raise urgent ethical concerns about privacy and misinformation," and the authors state the models' "ability to reproduce social complexity remains uneven and unproven." Validation, they write, is "the central challenge for generative social simulation."

The framing is self-referential: "As artificial intelligence reshapes human society, understanding this transformation requires the very tools driving it—a recursive challenge that will define the future of social science." Two researchers we track posted the link within days of publication, a small signal that the taxonomy is already circulating inside the field.