Out of One, Many: Using Language Models to Simulate Human Samples

TL;DR

This paper demonstrates that GPT-3 can be conditioned to accurately emulate diverse human subgroups, providing a nuanced tool for social science research by creating detailed 'silicon samples' that reflect complex human attitudes and demographics.

Contribution

The study introduces the concept of 'algorithmic fidelity' in language models, showing they can serve as effective proxies for human subpopulations in social science research.

Findings

GPT-3's biases are demographically correlated and can be fine-tuned.

Silicon samples closely match real human survey responses.

Language models capture complex socio-cultural attitudes.

Abstract

We propose and explore the possibility that language models can be studied as effective proxies for specific human sub-populations in social science research. Practical and research applications of artificial intelligence tools have sometimes been limited by problematic biases (such as racism or sexism), which are often treated as uniform properties of the models. We show that the "algorithmic bias" within one such tool -- the GPT-3 language model -- is instead both fine-grained and demographically correlated, meaning that proper conditioning will cause it to accurately emulate response distributions from a wide variety of human subgroups. We term this property "algorithmic fidelity" and explore its extent in GPT-3. We create "silicon samples" by conditioning the model on thousands of socio-demographic backstories from real human participants in multiple large surveys conducted in the United States. We then compare the silicon and human samples to demonstrate that the information contained in GPT-3 goes far beyond surface similarity. It is nuanced, multifaceted, and reflects the complex interplay between ideas, attitudes, and socio-cultural context that characterize human attitudes. We suggest that language models with sufficient algorithmic fidelity thus constitute a novel and powerful tool to advance understanding of humans and society across a variety of disciplines.