Generative AI adoption and employee empowerment in tourism: ChatGPT use for text-based customer communication and its effects on workplace power dynamics

Abstract

This study investigates tourism and hospitality employees' behavioral intentions to adopt ChatGPT for text-based customer communication and examines how generative AI use shapes psychological empowerment and internal power dynamics within service teams. Drawing on a mixed-methods design that integrates survey data from 322 employees in Sri Lanka's tourism sector with qualitative thematic analysis, the study applies an integrated framework combining the Technology Acceptance Model (TAM) and the Theory of Planned Behavior (TPB). PLS-SEM results indicate that perceived usefulness and perceived ease of use positively influence employee attitude toward ChatGPT, while attitude, subjective norm, and perceived behavioral control together account for 64.3% of the variance in intention to use. The integrated model demonstrates strong explanatory power, with attitude emerging as the dominant predictor of adoption intention. Qualitative findings reveal that ChatGPT use enhances psychological empowerment across four dimensions: meaning, self-determination, competence, and impact. Critically, increased employee autonomy in communication tasks is accompanied by a redistribution of communicative authority within teams, shifting control from supervisory gatekeepers toward frontline and back-office staff. These results establish that generative AI functions not merely as a productivity instrument but as a structural force reshaping work roles, decision-making authority, and organizational power relations in tourism settings. The study contributes novel empirical evidence at the intersection of AI adoption theory, empowerment theory, and power-agency perspectives in the context of an emerging economy's tourism sector.