Algorithm Aversion: Evidence from Ridesharing Drivers.

This article examines algorithm aversion among drivers on a large Asian ridesharing platform, focusing on why drivers often resist following AI-based repositioning recommendations designed to optimize system-wide efficiency. The study identifies two key factors influencing drivers' reluctance: their contextual experience—drivers are less likely to follow recommendations that conflict with their own past location-time experiences—and herding behavior, where drivers tend to imitate peers' location choices, especially under uncertainty. Using detailed task-level data and driver surveys, the research finds that greater driver experience and positive algorithm outcomes reduce herding effects and increase algorithm adoption. The findings highlight challenges in aligning individual incentives with system-wide objectives and suggest that firms should tailor communication and incentives based on users' contextual experience and peer influences to improve algorithm uptake.