Can Generative AI Improve Hotel Service Recovery? Employee Trust and Adoption of AI- Assisted Decision Support

Generative artificial intelligence (GenAI) is increasingly capable of supporting frontline employees in handling complex service failures, yet its role in hotel service recovery remains insufficiently understood. Unlike routine service tasks, service recovery requires employees to interpret guest complaints, evaluate alternative responses, consider compensation options, and make timely decisions under operational constraints. This study examines how hotel employees evaluate and adopt GenAI-assisted decision support for service recovery. Drawing on human–AI trust and task–technology fit perspectives, the proposed model investigates the effects of AI response quality and AI explainability on trust in AI, as well as the influence of task–AI fit on perceived decision support. Trust in AI and perceived decision support are subsequently examined as predictors of employees’ adoption intention, which is expected to enhance perceived service recovery decision effectiveness. A quantitative, scenario-based survey will be conducted among hotel employees with experience in handling guest complaints and service recovery situations. Data will be analyzed using partial least squares structural equation modeling (PLS-SEM). The study extends existing research beyond general GenAI adoption by focusing on a specific, judgment-intensive hospitality task and positioning GenAI as a human-in-the-loop decision-support mechanism rather than an autonomous decision maker. The findings are expected to provide insights into the technological and cognitive conditions under which hotel employees are willing to rely on GenAI recommendations to support more informed, consistent, and effective service recovery decisions.