{"title":"Generative AI vs. humans in online hotel review management: A Task-Technology Fit perspective","authors":"Huihui Zhang , Zheng Xiang , Florian J. Zach","doi":"10.1016/j.tourman.2025.105187","DOIUrl":null,"url":null,"abstract":"<div><div>Despite Generative AI's ability to produce human-like content, its effectiveness as references for human responses, particularly in online review management, remains unclear. To address this question, this study explores if human responses resembling AI patterns are associated with enhanced customer perceptions. To provide deeper insights, we examined how this relationship shifts under varying technological and task conditions, guided by the Task-Technology Fit theory. In the empirical analysis, we automated responses to 32,129 online reviews using GPT, calculated the similarity between existing managerial responses and AI-generated content, and tested the relationship between human-AI similarity and the perceived helpfulness of review-response pairs. The findings reveal benefits of resembling AI with high model temperatures, particularly for positive reviews, while identifying negative outcomes under lower temperatures. This study enriches our understanding of an emerging technology that could have a huge impact on the industry and provides insights for practitioners to refine AI adoption strategies.</div></div>","PeriodicalId":48469,"journal":{"name":"Tourism Management","volume":"110 ","pages":"Article 105187"},"PeriodicalIF":10.9000,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Tourism Management","FirstCategoryId":"91","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0261517725000573","RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL STUDIES","Score":null,"Total":0}
引用次数: 0
Abstract
Despite Generative AI's ability to produce human-like content, its effectiveness as references for human responses, particularly in online review management, remains unclear. To address this question, this study explores if human responses resembling AI patterns are associated with enhanced customer perceptions. To provide deeper insights, we examined how this relationship shifts under varying technological and task conditions, guided by the Task-Technology Fit theory. In the empirical analysis, we automated responses to 32,129 online reviews using GPT, calculated the similarity between existing managerial responses and AI-generated content, and tested the relationship between human-AI similarity and the perceived helpfulness of review-response pairs. The findings reveal benefits of resembling AI with high model temperatures, particularly for positive reviews, while identifying negative outcomes under lower temperatures. This study enriches our understanding of an emerging technology that could have a huge impact on the industry and provides insights for practitioners to refine AI adoption strategies.
期刊介绍:
Tourism Management, the preeminent scholarly journal, concentrates on the comprehensive management aspects, encompassing planning and policy, within the realm of travel and tourism. Adopting an interdisciplinary perspective, the journal delves into international, national, and regional tourism, addressing various management challenges. Its content mirrors this integrative approach, featuring primary research articles, progress in tourism research, case studies, research notes, discussions on current issues, and book reviews. Emphasizing scholarly rigor, all published papers are expected to contribute to theoretical and/or methodological advancements while offering specific insights relevant to tourism management and policy.