Answers are grounded in this paper and cite their sources.
With the rise of machine learning (ML), consumers’ personal data is widely used for personalized product recommendations. To address privacy concerns, regulations increasingly grant consumers control over their data. One popular implementation is the “opt-out of information use” feature which allows consumers to specify which of their collected personal data recommender systems can harness. However, we conjecture that such features may have an unintended side effect: withholding data could inadvertently reveal insights about consumers’ latent characteristics, thereby enhancing targeting possibilities. Through a controlled experiment, we evaluate both consumers’ perceptions and technical consequences of such opt-out features in the context of a typical search problem for which we train an ML-based recommender. Our results show that these features increase consumers’ sense of control over the system and reduce perceived privacy intrusion among those who actually withhold information. Paradoxically, withholding information can simultaneously be harnessed to improve the ML model’s predictive accuracy. From a regulatory perspective, we highlight the need for specifying rules on how organizations may use information withholding decisions when consumers’ interests conflict with those of the recommender provider.
No takes yet. Share an insight, caveat, or question.
Zacharias et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: