Artificial intelligence and the reproduction of inequality: The role of algorithmic bias in social class imaginaries

Abstract

The expansion of artificial intelligence has intensified debates concerning its social consequences, yet prevailing perspectives often assume algorithmic neutrality. Existing research insufficiently theorizes how digital infrastructures reproduce class-based inequalities, particularly within education, labor markets, and access to financial resources. Addressing this gap, the present study develops a critical framework grounded in Pierre Bourdieu's concepts of reproduction and habitus, integrating insights from digital sociology to interrogate the structural effects of algorithmic systems. Employing document analysis and theoretical critique, the study examines how technological determinism obscures the translation of cultural, historical, and economic capital into computational processes. The findings suggest that artificial intelligence does not function as a neutral arbiter but rather consolidates existing power relations by embedding class-based preconceptions into digital decision-making. This dynamic risks crystallizing social stratification into what may be conceptualized as a digital caste system, thereby constraining social mobility and undermining democratic publicness. The study concludes that overcoming the crisis of legitimacy surrounding algorithmic governance requires abandoning the illusion of neutrality and advancing a data justice framework capable of monitoring and transforming the class consequences of algorithmic rule.

Keywords: Algorithmic governance; data justice; digital inequality; habitus; social reproduction.