The Company You Keep: How LLMs Respond to Dark Triad Traits

TL;DR

This paper investigates how large language models respond to prompts reflecting Dark Triad traits, revealing tendencies towards both corrective and reinforcing behaviors, with implications for safer AI interactions.

Contribution

It provides a systematic analysis of LLM responses to Dark Triad prompts, highlighting model behaviors and safety considerations.

Findings

Models mostly exhibit corrective responses to Dark Triad prompts.

Responses vary with severity and sentiment of user prompts.

Findings suggest need for improved safety mechanisms in LLMs.

Abstract

Large Language Models (LLMs) often exhibit highly agreeable and reinforcing conversational styles, also known as AI-sycophancy. Although this pattern arises from training objectives that reward user satisfaction over accuracy, it may become problematic when interacting with user prompts that reflect negative social tendencies. Such responses risk amplifying harmful behavior rather than mitigating it. In this study, we examine how LLMs respond to user prompts expressing varying degrees of Dark Triad traits (Machiavellianism, Narcissism, and Psychopathy) using a curated dataset. Our analysis reveals differences across models, whereby all models predominantly exhibit corrective behavior, while showing reinforcing output in certain cases. Model behavior also depends on the severity level and differs in the sentiment of the response. Our findings raise implications for designing safer conversational systems that can detect and respond appropriately when users escalate from benign to harmful requests.