What Drives Tourists to Use Conversational AI? Evidence from an Extended UTAUT2 Model

Abstract

This study examines the factors shaping tourists’ adoption and use of artificial intelligence (AI) tools in tourism by drawing on the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2). Focusing on tourists visiting Fes, Morocco, AI is operationalised as a conversational assistant similar to ChatGPT that helps travellers search for information, plan itineraries, and make decisions during their trip. The research investigates how performance expectancy, effort expectancy, social influence, facilitating conditions, hedonic motivation, price value, and habit influence behavioural intention and usage behaviour, and whether trust in AI technology moderates the intention–behaviour link. Data were collected through an online survey of N=183 tourists in Fes, Morocco. Respondents completed a screening question on prior use of a ChatGPT-like assistant for travel tasks. Only “Yes” respondents proceeded. Usage behavior reflects general self-reported use of AI for travel purposes, not verified frequency of use during the Fes stay. The collected data was then analyzed using a consistent “partial least squares” (PLS) SEM technique. The findings show that performance expectancy, hedonic motivation, price value, and habit significantly predict behavioural intention. Facilitating conditions and habit significantly predict usage behaviour. Effort expectancy and social influence are non-significant predictors of intention. Trust in AI technology does not significantly moderate the relationship between intention and usage behaviour. The study contributes to tourism technology literature by refining the application of UTAUT2 in an AI-intensive context and highlighting the role of habit and perceived value in tourists’ continued reliance on AI assistants. Practical implications are discussed for tourism managers and destination stakeholders seeking to design and communicate trustworthy AI-based services.