The Personality Trap: How LLMs Embed Bias When Generating Human-Like Personas

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Amidei, Jacopo, Ferreira, Gregorio, Serrano, Mario Muñoz, Nieto, Rubén, Kaltenbrunner, Andreas
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915770517159936
author Amidei, Jacopo
Ferreira, Gregorio
Serrano, Mario Muñoz
Nieto, Rubén
Kaltenbrunner, Andreas
author_facet Amidei, Jacopo
Ferreira, Gregorio
Serrano, Mario Muñoz
Nieto, Rubén
Kaltenbrunner, Andreas
contents This paper examines biases in large language models (LLMs) when generating synthetic populations from responses to personality questionnaires. Using five LLMs, we first assess the representativeness and potential biases in the sociodemographic attributes of the generated personas, as well as their alignment with the intended personality traits. While LLMs successfully reproduce known correlations between personality and sociodemographic variables, all models exhibit pronounced WEIRD (western, educated, industrialized, rich and democratic) biases, favoring young, educated, white, heterosexual, Western individuals with centrist or progressive political views and secular or Christian beliefs. In a second analysis, we manipulate input traits to maximize Neuroticism and Psychoticism scores. Notably, when Psychoticism is maximized, several models produce an overrepresentation of non-binary and LGBTQ+ identities, raising concerns about stereotyping and the potential pathologization of marginalized groups. Our findings highlight both the potential and the risks of using LLMs to generate psychologically grounded synthetic populations.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03334
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Personality Trap: How LLMs Embed Bias When Generating Human-Like Personas
Amidei, Jacopo
Ferreira, Gregorio
Serrano, Mario Muñoz
Nieto, Rubén
Kaltenbrunner, Andreas
Computers and Society
This paper examines biases in large language models (LLMs) when generating synthetic populations from responses to personality questionnaires. Using five LLMs, we first assess the representativeness and potential biases in the sociodemographic attributes of the generated personas, as well as their alignment with the intended personality traits. While LLMs successfully reproduce known correlations between personality and sociodemographic variables, all models exhibit pronounced WEIRD (western, educated, industrialized, rich and democratic) biases, favoring young, educated, white, heterosexual, Western individuals with centrist or progressive political views and secular or Christian beliefs. In a second analysis, we manipulate input traits to maximize Neuroticism and Psychoticism scores. Notably, when Psychoticism is maximized, several models produce an overrepresentation of non-binary and LGBTQ+ identities, raising concerns about stereotyping and the potential pathologization of marginalized groups. Our findings highlight both the potential and the risks of using LLMs to generate psychologically grounded synthetic populations.
title The Personality Trap: How LLMs Embed Bias When Generating Human-Like Personas
topic Computers and Society
url https://arxiv.org/abs/2602.03334