Service Interactions and Gig Workers Safety Behavior: An Action Identification Perspective

While the gig economy reshapes workers’ meaning-making processes, research has yet to fully capture the dynamic shifts in meaning that underpin their daily work experience. Drawing from action identification theory, we examine how human versus algorithm-mediated service interactions affect gig workers’ identification with their work and subsequent safety behaviors. Using the experience sampling method to collect data from delivery riders, our results reveal that positive human service interactions increase workplace identification and safety behaviors, whereas negative algorithm-mediated service interaction shifts identification towards bottom-line outcomes, leading to increased dangerous behaviors, particularly among workers facing high relational challenges. Overall, our research contributes to the understanding of gig workers’ meaning construction process under algorithm management during their daily customer service experience and offers practical implications to promote gig work safety.