Topic: dark triad

The study explores how large language models (LLMs) are influenced by social desirability and impression management, similar to humans during personality assessments, highlighting the need for better understanding of response distortions in AI.

arxiv.org↗