<![CDATA[The digital transformation of public services has produced a policy paradox: services designed to ease citizen interaction with the state often introduce new forms of administrative burden. This narrative literature review examines the role of Artificial Intelligence (AI) in reducing administrative burden in digital public services, using Herd and Moynihan's theoretical framework that defines administrative burden through three dimensions: learning costs, compliance costs, and psychological costs. The synthesis covers 28 Scopus- and SINTA-indexed publications from 2015 to 2025. The narrative synthesis yields three main findings: first, AI through chatbots, document automation, and intelligent systems can substantially reduce compliance and learning costs by enabling 24/7 access, personalisation, and procedural navigation. Second, AI's impact on psychological costs is dualistic and moderated by trust, digital literacy, and user experience (UX) design. Third, in developing countries such as Indonesia, infrastructure gaps and the digital divide can reverse AI's benefits into additional burdens for vulnerable groups. This article maps four significant research gaps theoretical, empirical-geographical, methodological, and contextual (generative AI) and proposes a future research agenda for advancing public administration scholarship in the age of AI.]]>
Copyrights © 2026
