Be Friendly, Not Friends: How LLM Sycophancy Shapes User Trust

TL;DR

This study investigates how different styles of LLM responses influence user trust, revealing that conversational demeanor and stance adaptation significantly affect perceptions of authenticity and trustworthiness.

Contribution

It introduces a user-centric framework for understanding LLM sycophancy and demonstrates how response styles impact user trust through experimental evidence.

Findings

Complimentary, adaptive LLMs reduce perceived authenticity and trust.

Neutral, adaptive LLMs increase perceived authenticity and trust.

Response style manipulation can influence over-trusting behaviors.

Abstract

LLM-powered conversational agents are increasingly influencing our decision-making, raising concerns about "sycophancy" - the tendency for LLMs to excessively agree with users even at the expense of truthfulness. While prior work has primarily examined LLM sycophancy as a model behavior, our understanding of how users perceive this phenomenon and its impact on user trust remains significantly lacking. In this work, we conceptualize LLM sycophancy along two key constructs: conversational demeanor (complimentary vs. neutral) and stance adaptation (adaptive vs. consistent). A 2 x 2 between-subjects experiment (N = 224) revealed complex dynamics: complimentary LLMs that adapted their stance reduced perceived authenticity and trust, while neutral LLMs that adapted enhanced both, suggesting a pathway for manipulating users into over-trusting LLMs beyond their actual capabilities. Our findings advance user-centric understanding of LLM sycophancy and provide profound implications for developing more ethical and trustworthy LLM systems.