The Effects of Self-Disclosing in Writing with AI Writing Assistants (vs. Without) on Well-Being
(A2025-125702)

AUTHORS

Atil Mansour, Technion, Israel Institute of Technology; Ofra Amir, Technion, Israel Institute of Technology; Liat Levontin, Technion, Israel Institute of Technology

ABSTRACT

Using language to reveal ourselves is a process called self-disclosure. Self-disclosure is important as it was found to affect well-being positively. Consumers often self-disclose while writing about personal experiences when writing reviews and social media posts. Recently, there has been an increase in AI-mediated communication, such that LLM-based conversational agents are integrated into critical application areas and can affect consumers’ language and thought. The current research tested the effects of self-disclosing in writing while using AI (vs. without) on well-being. We hypothesized that self-disclosing in writing using AI (vs. no use) would be negatively related to well-being, leading to a lower positive and a higher negative affect. Although we found no such differences between the conditions, text analysis revealed that texts containing AI's words (vs. not) were indeed less self-disclosing.