Algorithmic Discrimination: The Making of Precarity in Platform Work

• In Event: 30606 - Exploring the Impacts of Platform Work

Abstract

This study examines the production of precarity in the platform economy, focusing on how algorithmic discrimination structures the experiences and dispositions of gig workers. While prior research has debated the universality of precarity in platform work, this study argues that precarity is differentially operationalized through algorithmic management systems that sort and match workers based on performance ratings. Drawing on a year of ethnographic fieldwork as a temporary laborer for an online staffing platform, this research explores how workers contest and navigate algorithmic sorting mechanisms. Ethnographic findings reveal that workers engage in "algorithmic performances"—strategic behaviors aimed at maximizing visibility to rating systems—to secure more stable and higher-paying work. Building on these insights, an agent-based model (ABM) simulates platform labor dynamics, demonstrating how algorithmic discrimination leads to self-reinforcing cycles of worker stratification, where those with lower initial ratings face persistent disadvantage and income instability. The model further highlights how employer biases embedded in rating systems exacerbate structural inequality, creating 'lock-in' effects for precarized workers. This study contributes to scholarship on digital labor by conceptualizing algorithmic discrimination as a distinct mode of labor control and by illustrating how workers actively negotiate their conditions within algorithmic governance frameworks.

Author

• Vivek Ramakrishnan, University of California-Los Angeles