Public Trust in Government Chatbots for Administrative Services

Abstract

Background: The practical operation of policy relating to government chatbots differs across settings, suggesting a need to examine implementation capacity and institutional design directly.

Objective: This study sought to explain differences in the implementation of government chatbots by examining institutional responsibility, administrative capacity, and public trust.

Methods: A cross-sectional randomized survey experiment was designed with 480 adult participants. Respondents were allocated to alternative policy-information conditions, after which perceived legitimacy, understanding, trust, and intended compliance were measured using standardized scales. Group differences were assessed using regression models adjusted for demographic and institutional covariates.

Results: The clearer communication condition increased both understanding and trust. Relative to the comparison group, the adjusted difference on the primary outcome was 0.24 standard deviations and intended compliance rose by 17%. Gains were largest among participants who began with less familiarity with the policy area.

Conclusion: The findings underline a practical gap between formal policy on government chatbots and the institutional work required to deliver it consistently. This means that public trust and participation depend on transparent communication, accessible procedures, and visible accountability for implementation.

Study design: citizen survey experiment.

Publication history: Originally published in print on 15 June 2025, Volume 5, Article 10. Online abstract record added on 16 September 2026. Full-text PDF will be added when the digital galley is ready.