Misalignment of LLM-Generated Personas with Human Perceptions in Low-Resource Settings

TL;DR

This paper investigates how well LLM-generated personas match human perceptions in Bangladesh, revealing significant gaps in empathy, credibility, and cultural understanding, highlighting the need for validation against real-world data.

Contribution

It provides a quantitative comparison of LLM-generated personas with human responses in a low-resource setting, emphasizing the importance of alignment validation.

Findings

Humans outperform LLMs in response accuracy and perception matrices.

LLMs exhibit systematic positivity bias along the Pollyanna Principle.

Significant gaps in empathy and credibility between LLMs and humans.

Abstract

Recent advances enable Large Language Models (LLMs) to generate AI personas, yet their lack of deep contextual, cultural, and emotional understanding poses a significant limitation. This study quantitatively compared human responses with those of eight LLM-generated social personas (e.g., Male, Female, Muslim, Political Supporter) within a low-resource environment like Bangladesh, using culturally specific questions. Results show human responses significantly outperform all LLMs in answering questions, and across all matrices of persona perception, with particularly large gaps in empathy and credibility. Furthermore, LLM-generated content exhibited a systematic bias along the lines of the ``Pollyanna Principle'', scoring measurably higher in positive sentiment ($\Phi_{avg} = 5.99$ for LLMs vs. $5.60$ for Humans). These findings suggest that LLM personas do not accurately reflect the authentic experience of real people in resource-scarce environments. It is essential to validate LLM personas against real-world human data to ensure their alignment and reliability before deploying them in social science research.