Citation

Frough, Eid Mohammad and Hassan, Hasliza and Nathan, Robert Jeyakumar (2026) AI chatbots in the food and beverage industry: Trust, privacy calculus, and customer engagement in a developing economy. Computers in Human Behavior Reports, 23. p. 101274. ISSN 24519588

Text AI chatbots in the food and beverage industry_ Trust, privacy calculus, and customer engagement in a developing economy.pdf - Published Version Restricted to Repository staff only Download (5MB)

Abstract

AI-powered chatbots are rapidly transforming service interactions; however, their adoption within small and medium-sized enterprises (SMEs) in developing economies remains uneven. In Malaysia's food and beverage (F&B) sector, where speed, personalisation, and service reliability are central to customer experience, chatbot uptake is notably low, driven largely by consumer uncertainty about privacy and trust. This research is conducted to advance the theoretical and practical understanding of this gap. The study integrates the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) with the Privacy Calculus to model the cognitive, affective, experiential, and risk–benefit evaluations that shape chatbot adoption as a sequential, trust-dependent process. Through a quantitative cross-sectional research design, the study collected responses from 595 Malaysian caf´e customers. Employing PLS-SEM, the findings demonstrate that performance expectancy, effort expectancy, hedonic motivation, and habit substantially strengthen behavioural intention, while intention and habit jointly are associated with actual use. Trust emerges as a decisive mechanism: perceived benefits strongly enhance trust, whereas perceived privacy risk markedly erodes it. Trust is further associated with customer satisfaction both directly and through behavioural intention and actual use. Moderation analyses reveal that age shapes the influence of hedonic motivation, and experience modifies reliance on facilitating conditions. The study contributes to AI-adoption theory in three specific ways: it positions trust as the empirically tested mediating mechanism through which the Privacy Calculus and UTAUT2 are integrated rather than additively combined; it extends the chatbot-adoption evidence base to the under-represented context of Malaysian caf´e and coffee-shop SMEs in a developing economy; and it operationalises chatbot engagement as an ongoing trustdependent sequence from privacy-benefit evaluation through intention, actual use, and satisfaction, rather than a one-shot acceptance decision.

Item Type:	Article
Uncontrolled Keywords:	SMEs Malaysia
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD2321-4730.9 Industry > HD2340.8-2346.5 Small and medium-sized businesses, artisans, handicrafts, trades
Divisions:	Faculty of Management (FOM)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	02 Sep 2026 08:18
Last Modified:	02 Sep 2026 08:18
URII:	http://shdl.mmu.edu.my/id/eprint/16546

Downloads

Edit (login required)