Human-machine communication privacy management, privacy fatigue, and the conditional effects of algorithm awareness on privacy co-ownership in the social media context

Abstract

Data about individual users drives today's social media content-filtering algorithm recommendations. Through nuanced interactions with social media algorithms, such as human-algorithm interplay, the end user effortlessly cultivates a social media feed. While this level of personalization can significantly benefit the user, recommended ads and content sometimes resemble aspects of the user's private lives that they may not have wanted the algorithm or platform to know. Moreover, though users dislike these experiences of privacy violations, they still disclose private information to the system due to fatigue in managing online privacy altogether. This current study integrates communication privacy management (CPM) theory (Petronio, 2002) into the human-algorithm interaction context to examine the extent to which social media users (N = 1,305) engage in open privacy management practices with social media platforms via their algorithms, depending on their felt privacy fatigue. Results from using latent moderated structural equations (LMS) suggest that individuals' awareness of algorithms is negatively associated with using open privacy management practices with social media algorithms. However, this depends on their felt privacy fatigue, such that individuals who are both highly aware and highly fatigued are likely to be more closed off in sharing private information with social media algorithms, thus granting less co-ownership rights to social media platforms. In light of these findings, implications for future research on communication privacy management in the context of social media algorithms are discussed.