Paper Number

ICIS2025-2125

Paper Type

Short

Abstract

On-demand service platforms using a postpaid model often encounter delayed payment challenges due to the absence of credit obligations. SMS-based reminders that involve personal information are widely employed to urge users to pay. However, this practice creates a privacy-efficiency paradox, prompting platforms to seek ways to facilitate collections while safeguarding user privacy. Drawing on Contextual Integrity Theory, we investigate how sender identity (AI vs. human) and information types (personally identifiable, non-personally identifiable, or no information) in SMS reminders influence collection effectiveness. Our randomized field experiment reveals that the combination of AI sender and non-personally identifiable information results in the highest repayment rates. This phenomenon may be explained by a balance between lower perceived intrusion and retained deterrence. Our findings contribute to research on the privacy paradox, human-AI perception asymmetries, and debt collection, providing insights for designing effective, privacy-compliant reminders.

Comments

09-Cybersecurity

Share

COinS
 
Dec 14th, 12:00 AM

When is AI Superior to Human? Unveiling the Effects of “Word-of-Machine” on Debt Collection Utilizing Different Types of Privacy

On-demand service platforms using a postpaid model often encounter delayed payment challenges due to the absence of credit obligations. SMS-based reminders that involve personal information are widely employed to urge users to pay. However, this practice creates a privacy-efficiency paradox, prompting platforms to seek ways to facilitate collections while safeguarding user privacy. Drawing on Contextual Integrity Theory, we investigate how sender identity (AI vs. human) and information types (personally identifiable, non-personally identifiable, or no information) in SMS reminders influence collection effectiveness. Our randomized field experiment reveals that the combination of AI sender and non-personally identifiable information results in the highest repayment rates. This phenomenon may be explained by a balance between lower perceived intrusion and retained deterrence. Our findings contribute to research on the privacy paradox, human-AI perception asymmetries, and debt collection, providing insights for designing effective, privacy-compliant reminders.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.