The Illusion of Agreement with ChatGPT: Sycophancy and Beyond

TL;DR

This paper analyzes Reddit discussions to explore user-reported harms from ChatGPT, such as delusion and addiction, and suggests multi-level strategies for mitigation involving users, developers, and policymakers.

Contribution

It provides a novel analysis of user experiences and concerns regarding ChatGPT harms, highlighting the need for coordinated mitigation strategies.

Findings

Identified five user-reported concerns across various life domains.

Documented three levels of user-driven mitigation strategies.

Highlighted the importance of multi-stakeholder interventions.

Abstract

While concerns about ChatGPT-induced harms due to sycophancy and other behaviors, including gaslighting, have grown among researchers, how users themselves experience and mitigate these harms remain largely underexplored. We analyze Reddit discussions to investigate what concerns users report and how they address them. Our findings reveal five distinct user-reported concerns that manifest across multiple life domains, ranging from personal to societal: inducing delusion, digressing narratives, implicating users for models' limitations, inducing addiction, and providing unsupervised psychological support. We document three-tier user-driven suggestions spanning functional usage techniques, behavioral approaches, and private and institutional safeguards. Our findings show that AI-induced harms require coordinated interventions across users, developers, and policymakers. We discuss design implications and future directions to mitigate the harms and ensure user benefits.