Nature Human Behaviour review maps LLM roles in social science

A new Review in Nature Human Behaviour sorts the way large language models are being used across social science into four distinct roles: "subjects that simulate human behaviour, assistants that automate analysis, collaborators that generate hypotheses and evaluators that audit and improve research quality."

The scope is wide. The authors survey work across sociology, political science, psychology, anthropology and economics, and describe LLMs as "both our research tools and authentic research subjects."

The Review is blunt about where the technology falls short. "LLMs struggle with causal inference and raise urgent ethical concerns about privacy and misinformation, and their ability to reproduce social complexity remains uneven and unproven," the abstract states. The authors call this "a recursive challenge that will define the future of social science."

The abstract publishes no headline benchmarks for any of the four roles. It names "critical research directions" and recent advances rather than a scoreboard.

Two researchers we follow shared the paper on publication.