LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users
Abstract
While state-of-the-art large language models (LLMs) have shown impressive performance on many tasks, systematically evaluating undesirable behaviors of these models remains critical. In this work, we investigate how the quality of LLM responses changes in terms of information accuracy, truthfulness, and refusals depending on three user traits: English proficiency, education level, and country of origin. We present extensive experimentation on three state-of-the-art LLMs and two different datasets targeting truthfulness and factuality. Our findings suggest that undesirable behaviors in state-of-the-art LLMs occur disproportionately more for users with lower English proficiency, of lower education status, and originating from outside the US, rendering these models unreliable sources of information towards their most vulnerable users.
Fields of Study & Subjects
Topic Modeling
Artificial Intelligence in Healthcare and Education
Misinformation and Its Impacts
Publication details
- Journal / source
- Proceedings of the AAAI Conference on Artificial Intelligence
- Publisher
- Association for the Advancement of Artificial Intelligence (AAAI)
- Volume
- 40
- Issue
- 46
- Pages
- 39116–39124
- Publication date
- 14 March 2026