The Power of AI-Generated Content: Evidence From the Peer-to-Peer Accommodation Market

被引:1
作者
Fan, Ningyuan [1 ]
Li, Xiang [2 ]
Liu, Chao [3 ]
Fan, Zhi-Ping [4 ,5 ]
机构
[1] Nankai Univ, Coll Tourism & Serv Management, Tianjin, Peoples R China
[2] Temple Univ, Sch Sport Tourism & Hospitality Management, Philadelphia, PA USA
[3] City Univ Hong Kong, Dept Informat Syst, Hong Kong, Peoples R China
[4] Northeastern Univ, Sch Business Adm, Shenyang, Peoples R China
[5] Northeastern Univ, Natl Frontiers Sci Ctr Ind Intelligence & Syst Opt, Shenyang, Peoples R China
基金
中国国家自然科学基金;
关键词
artificial intelligence; generative AI; AIGC; prompt engineering; GPT-4o; peer-to-peer accommodation market; SIGNALING THEORY; SHARING ECONOMY; AIRBNB; DESTINATION; ATTRIBUTES; ALGORITHMS; QUALITY;
D O I
10.1177/00472875251332951
中图分类号
F [经济];
学科分类号
02 ;
摘要
As a technological breakthrough, large language models stand to revolutionize operations in many industries, including the tourism sector. Despite the transformative potential of artificial intelligence (AI)-generated content (AIGC), its role in marketing performance remains unclear. This research used a prediction-interpretation-inference machine learning framework to evaluate how AIGC affects Airbnb listings' performance. Specifically, we scrutinized AIGC's efficacy under two conditions (i.e., with an AIGC tool serving as a language assistant vs. a content creator) and examined the impacts of prompt design strategies. Three key findings emerged. First, the prompt design strategy greatly influenced AIGC's effectiveness. Second, AIGC did not uniformly enhance listing performance: although it boosted revenue and bookings for underperforming listings, its impact on low-performing listings was limited and not as strong as anticipated. AIGC may even negatively influence high-performing listings. Third, the AIGC tool appeared more effective as a language assistant (vs. a content creator) for underperforming listings.
引用
收藏
页数:20
相关论文
共 105 条
[1]   Transformer models for text-based emotion detection: a review of BERT-based approaches [J].
Acheampong, Francisca Adoma ;
Nunoo-Mensah, Henry ;
Chen, Wenyu .
ARTIFICIAL INTELLIGENCE REVIEW, 2021, 54 (08) :5789-5829
[2]  
Aira Verblio, 2022, The future of AI in content marketing
[3]   Visualizing the effects of predictor variables in black box supervised learning models [J].
Apley, Daniel W. ;
Zhu, Jingyu .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2020, 82 (04) :1059-1086
[4]   A CONSUMER-SIDE EXPERIMENTAL EXAMINATION OF SIGNALING THEORY - DO CONSUMERS PERCEIVE WARRANTIES AS SIGNALS OF QUALITY [J].
BOULDING, W ;
KIRMANI, A .
JOURNAL OF CONSUMER RESEARCH, 1993, 20 (01) :111-123
[5]   Tourism brochures: Linking message strategies, tactics and brand destination attributes [J].
Brito, Pedro Quelhas ;
Pratas, Joaquim .
TOURISM MANAGEMENT, 2015, 48 :123-138
[6]  
Cao YH, 2023, Arxiv, DOI [arXiv:2303.04226, 10.48550/arXiv.2303.04226, DOI 10.48550/ARXIV.2303.04226]
[7]   ChatGPT for tourism: applications, benefits and risks [J].
Carvalho, Ines ;
Ivanov, Stanislav .
TOURISM REVIEW, 2024, 79 (02) :290-303
[8]   How Closely Do Hypothetical Surveys and Laboratory Experiments Predict Field Behavior? [J].
Chang, Jae Bong ;
Lusk, Jayson L. ;
Norwood, F. Bailey .
AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS, 2009, 91 (02) :518-534
[9]   Do airbnb host listing attributes influence room pricing homogenously? [J].
Chattopadhyay, Manojit ;
Mitra, Subrata Kumar .
INTERNATIONAL JOURNAL OF HOSPITALITY MANAGEMENT, 2019, 81 :54-64
[10]   A linguistic signaling model of social support exchange in online health communities [J].
Chen, Langtao ;
Baird, Aaron ;
Straub, Detmar .
DECISION SUPPORT SYSTEMS, 2020, 130