Large language models can predict the results of social science experiments

Source: https://www.nature.com/articles/s41586-026-10742-x

Large language models can be used to estimate the results of social science experiments about as accurately as a group of human forecasters—even for experiments published after the models were trained—although they tended to overestim…

<li>Bail, C. A. Can generative AI improve social science? Proc. Natl Acad. Sci. USA121, e2314021121 (2024). Article  CAS  PubMed  PubMed Central  Google Scholar  </li><li>Grossmann, I. et al… [+13270 chars]