Exploring Consumer Avoidance Of AI Algorithm Recommendations On Short Video Platforms: A Mixed-Method Study In Vietnam

Abstract

Purpose - This study investigates how greedy and biased recommendations eventually increases consumers’ avoidance of products and services that AI algorithms recommended on short video platforms, with a focus on the role of technology intrusiveness, affiliate social media distrust, and perceived misinformation. Methodology - We employed a exploratory sequential mixed-method design for the research, combining both qualitative and quantitative methods. Initially, quantitative data was collected via online questionnaire with a sample of 533 young Vietnamese, analyzed by using the PLS-SEM model. Subsequently, qualitative data from 22 in-depth interviews was used to explain the quantitative findings and explore young users’ experience with AI recommenders on short video platforms. Findings - The findings reveal that greedy and biased recommendations increase technology intrusiveness, affiliate social media distrust, and perceived misinformation in users. The last two factors positively impact consumers’ avoidance, while the first factor has no significant impact. Technology intrusiveness wields positive influence on affiliate social media distrust. In addition, the qualitative research findings will be discussed in greater depth in the article. Originality/value - This is the first mixed method study to investigate more profound theoretical insights into consumers’ avoidance behavior on short video platforms and respond to the calls for research into the negative effects of AI on consumers with rigorous combined methods in the various digital environments. The study provides a comprehensive model which can help platform developers, policymakers, content creators, businesses and marketers to proactively formulate strategic responses to consumers on short video platforms with AI recommenders