Unintended Consequences of Recommender System Interventions: Evidence from a Field Experiment

2026-06-06

AI Summary

This study addresses the unintended long-term consequences of user interventions—such as “sleep reminders”—on recommender systems that dynamically adapt to user feedback. Through a large-scale field experiment on a short-video platform, combined with causal inference, log analysis, and dynamic policy modeling, the authors demonstrate that such interventions can inadvertently “retrain” the recommendation algorithm, inducing systemic shifts in content delivery. Contrary to expectations, the sleep reminder not only failed to reduce usage but increased late-night viewing duration by 14.75% and overall usage by 2.18%, with effects persisting for several weeks. These findings challenge the conventional paradigm of evaluating interventions under static assumptions and underscore the necessity of accounting for algorithmic adaptation in digital well-being policies.

Abstract

Platform content interventions in recommendation systems are typically evaluated as static "nudges", ignoring that the systems adaptively learn from the resulting user behavior. We investigate this dynamic through a large-scale field experiment on a short-video platform. The experiment involves a "sleep reminder" campaign designed to reduce late-night usage. Paradoxically, the intervention increased late-night engagement by 14.75% and overall platform usage by 2.18%, and the effects persisted for weeks even after the experiment. We explain this through a forced-exploration mechanism, showing that by revealing high latent demand for the promoted content, the intervention triggers a recommendation policy update that routine user behavior would not produce. The data generated by the intervention induced the algorithm to update its post-campaign policy, reinforcing the very engagement loops the campaign aimed to mitigate. Our findings demonstrate that user-facing interventions can effectively retrain the underlying algorithm, triggering durable, system-wide shifts in content distribution that challenge standard evaluation metrics in platform governance and social responsibility initiatives.

Problem

• recommender systems
• unintended consequences
• algorithmic adaptation
• platform interventions
• user behavior

Innovation

• recommender systems
• field experiment
• algorithmic adaptation
• forced exploration
• unintended consequences