Due to AI’s computation power, scalability, and cost efficiencies, algorithmic management (i.e., the managerial role played by AI) has become popular. Such mandated AI use provides benefits to the workers. This includes, but is not limited to, convenience, acquiring workplace skills, and improved productivity. However, mandated AI use also brings a critical challenge to the workers: the loss of human autonomy. While significant academic and practice debate has surrounded the AI use mandate, limited investigation has been conducted to uncover how such mandated AI use affects workers’ job performance and autonomy, as well as its impact on firm-level and societal level benefits.

In response to this gap, this project aims to measure the impact of mandated vs. autonomous AI use on empowerment and labor outcomes. We also aim to quantify the impact of mandated vs. autonomous AI use on platform-level and societal welfare (e.g., fairness and safety). To investigate these questions, we focus on the context of food delivery platforms which are at the forefront of developing and applying various AI applications as well as at the center of ongoing debates on the mandated use of AI. We obtained data from a randomised field experiment with two different modes of AI use (Human 100%, AI 100%, and Human-AI).