Large Language Models are overconfident and amplify human bias

Authors: F Sun, N Li, K Wang, L Goette

Published: 2025

Publication: arXiv preprint arXiv:2505.02151, 2025 - arxiv.org

Large language models (LLMs) exhibit overconfidence, amplifying human bias, especially in cases where their certainty declines, and their input doubles overconfidence in human decision making despite improving accuracy.

Methods: Algorithmically constructed reasoning problems with known ground truths were used to evaluate LLMs' confidence; comparisons were drawn with human performance using similar experimental protocols.

Key Findings: LLM confidence levels, correctness probabilities, comparison of bias between LLMs and humans, and effects of LLM input on human decision making.

Limitations: The study does not address how these biases may evolve in more diverse datasets or with further model improvements, nor does it explore strategies to mitigate LLM-induced overconfidence in humans.

Institution: HKU Business School

Research Area: LLM Overconfidence and Human Bias Amplification, Bias, LLM

Discipline: Artificial Intelligence, Behavioral Science

Citations: 21