UNDERSTANDING THE ADOPTION OF AI-BASED EMOTIONAL SUPPORT TOOLS: A UTAUT PERSPECTIVE ON CHATGPT USE AMONG EMPLOYEES IN THAILAND

Abstract

This study examines how employees adopt ChatGPT as an AI-enabled emotional support tool for mitigating burnout. Drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT), the study investigates the roles of performance expectancy, effort expectancy, social influence, and facilitating conditions in shaping adoption. A qualitative design was employed, based on semi-structured interviews with eight employees working in technology companies in Bangkok, Thailand. The findings show that performance expectancy was the strongest driver as participants valued ChatGPT for consultation, reflection, guidance, and encouragement. Effort expectancy supported adoption through conversational ease and immediacy, although prompting effort and voice-function interruptions sometimes reduced the quality of interaction. Social influence normalized trial through peers and social media, while facilitating conditions enabled continued use through either organizational access or personal initiative. Importantly, participants did not accept AI output uncritically; they filtered, adapted, and cross-checked advice before acting on it. Overall, ChatGPT was perceived as a supplementary rather than substitutive form of emotional support because users recognized risks such as impractical advice, over-reliance, and weakened human connection. The study extends UTAUT to an emotionally sensitive workplace context and identifies boundary conditions for the responsible use of AI-mediated support. Practically, the findings support bounded use of ChatGPT as a supplementary emotional-support tool, accompanied by user training, organizational guidance, and ethical design safeguards. However, the study is limited by a small qualitative sample from different firms; future research should use larger, longitudinal, and mixed-method designs across industries.

Keywords: ChatGPT, AI-enabled Emotional Support Tool, Artificial Intelligence, Thailand