Effects of distraction and anonymity on privacy trade-offs in facial recognition surveillance

Abstract

Despite the recent development of facial recognition technology (FRT) using AI, its implementation and applications are still controversial in countries. Amid the COVID-19 pandemic, FRT could be an effective crisis exit strategy for nations. However, there was a lack of empirical evidence in privacy literature or marketing research on how authorities, governments, and private sectors should implement the technology. Using the ground theory of ELM models, privacy calculus to examine privacy attitude, intention, and behavior, this study provides a theoretical model that illustrated people's thinking and decision-making process with FRT's applications. By conducting an online experiment with the random sampling of 603 respondents in the UK and the US, the study showed a structural model thinking process that led to privacy disclosure behavior. The findings confirmed that distraction and non-anonymized FRT would increase people's concerns about government intrusion, which mediately impact willingness to support and disclosure behavior. Moreover, moral consideration as moral equity referred that governments should raise awareness of FRT's importance to society to increase biometric data disclosure. Finally, a cluster analysis was conducted to classify people into three groups of "willing-to-share information." This further investigation also suggested a potential factor as stereotypes of people, which can deviate from the privacy paradox of people. This study contributed to privacy and marketing literatures in the context of FRT government surveillance.