The Algorithmic Anxiety Trap: Why A.I. Policy Must Move Beyond Post Work Welfare For India’s Gig Workers

Abstract

The rapid expansion of India's gig economy has accelerated the adoption of algorithmic management (AM), where artificial intelligence increasingly performs functions traditionally carried out by human supervisors. Although recent labour reforms, including the Social Security Code (2025), have expanded welfare provisions for gig workers, these measures continue to rely largely on post-employment protection and offer limited safeguards against the psychological consequences of AI-enabled workplace monitoring. As digital labour platforms become more dependent on automated decision-making, concerns surrounding transparency, continuous surveillance, and unexplained account deactivations have become increasingly prominent.

This study examines the Algorithmic Anxiety Trap, a condition in which opaque algorithmic systems and uncompensated digital labour contribute to sustained psychological distress among gig workers. Using a mixed-methods research design, the study investigates how platform governance influences worker well-being. The quantitative phase employs multiple linear regression analysis based on data collected from 100 respondents to examine the influence of Algorithmic Transparency and the Wage Gap Ratio, defined as the proportion of unpaid online waiting time relative to compensated working time. The qualitative phase complements these findings through Natural Language Processing (NLP)-based sentiment analysis of narrative responses from ten participants, enabling a deeper exploration of workers' emotional experiences and technology-related concerns.

The regression model demonstrates strong explanatory power (R²=0.823, p<0.001). Algorithmic Transparency is negatively associated with anxiety (β₁=-0.62), indicating that greater transparency reduces psychological distress. In contrast, the Wage Gap Ratio emerges as the strongest predictor of anxiety (β₂=2.00), suggesting that uncompensated periods of platform availability impose a substantial emotional burden on workers. The sentiment analysis reinforces these quantitative findings, producing a mean polarity score of −0.45 and revealing persistent concerns regarding automated grievance mechanisms, limited human intervention, and the anticipated impact of warehouse automation on future employment.

Taken together, the findings suggest that the psychological consequences of algorithmic management are shaped not only by technological opacity but also by the structural conditions under which platform work is organized. In particular, prolonged periods of unpaid availability appear to generate greater distress than algorithmic decision-making alone. The study therefore proposes a structural regulatory framework that incorporates availability pay into minimum wage calculations, establishes a legally enforceable right to human review of algorithmic decisions, and creates an Automation Transition Fund to support workers affected by technological displacement. These measures would strengthen labour protection while supporting a more equitable transition towards an AI-enabled economy.