ChatGPT for travel-related services: a pleasure–arousal–dominance perspective

Purpose This study uses the pleasure–arousal–dominance (PAD) theory to explore how users’ emotional engagement with ChatGPT drives their continued adoption of ChatGPT and word-of-mouth (WOM) behaviour in the context of travel-related service. Design/methodology/approach This study obtained reliable data from 428 Chinese respondents who used ChatGPT for travel-related purposes. Structural equation modelling was used to test a series of hypotheses based on the PAD framework. Findings This study identifies three key features of human–artificial intelligence (AI) interaction, namely, service ubiquity, entertainment and anthropomorphism, which significantly influence users’ emotional responses, including pleasure, arousal and dominance. Dominance and pleasure are found to enhance emotional experiences, driving continued adoption and positive WOM recommendations for ChatGPT, whereas arousal influences WOM but does not affect continued adoption. The results also confirm that users’ perceived pleasure from interacting with ChatGPT has the strongest effect. These findings advance theoretical understanding by clarifying the emotional mechanisms underlying human–AI interactions in the tourism context. Originality/value This study examines the emerging trend of tourists’ continuous adoption of ChatGPT for travel-related services. The results highlight how different emotions in human–AI interaction influence long-term use of AI-powered tool for travel-related services.