Nature paper finds LLMs can predict social science experiments with r=.85
RobbWiller · x · 2026-07-23
Paper overview
In Nature, the authors ask whether LLMs can predict the outcomes of social science experiments.
Evaluation setup
- 70 preregistered studies
- 469 experimental effects
- 119,330 participants
- GPT-4 was prompted with study materials and demographic profiles, then its simulated responses were used to estimate treatment effects
Result
The paper reports a strong correspondence of r = .85 between predicted and observed effects, suggesting LLMs may be useful for augmenting experimental social science.
Related event: Nature Study: GPT-4 Can Predict Social Science Experiment Results(18 posts)→
More from Research
- Meta FAIR's Self-Direction turns agent failures into reusable LoRA skills, +13.8pp on Terminal-Bench — karen_ullrich · 2026-10-27
- DONIC hits 86.79% on bimanual dexterous tracking, crushing prior SOTA's 9.43% — liuziwei7 · 2026-10-27
- Why don't more LLMs ship 10M-token context? A researcher asks about attention tradeoffs — agarwl_ · 2026-10-27
- PSNR below 15 dB in papers likely means the metric was misused, researcher warns — kwangmoo_yi · 2026-10-27
- Ortet's first health eval: 34 open-weight models, 50 tasks, generic rankings fall short — thegautamkamath · 2026-10-27
- Scaling-fit discrepancy: output-head FLOPs are ~65% of compute at the smallest rung, ~3% at 535B — dlwh · 2026-10-27