<![CDATA[Artificial intelligence (AI) is advancing more rapidly than the institutional arrangements designed to govern it, creating a persistent governance lag. Although agile governance and AI governance address this challenge from complementary perspectives, they have largely evolved as separate bodies of scholarship. This study develops an integrative conceptual framework by examining their convergence through a narrative review based on a transparent multi-source search and thematic synthesis. The analysis identifies three dimensions of convergence: diagnostic convergence, reflected in a shared recognition of governance lag and the Collingridge dilemma, normative convergence, demonstrated by common commitments to transparency, accountability, participation, and continuous learning, and instrumental convergence, expressed through regulatory sandboxes, policy laboratories, anticipatory governance, and lifecycle-based AI oversight. The findings further conceptualize agile AI governance as the application of agile governance meta-principles to AI governance while distinguishing it from adaptive, anticipatory, experimentalist, and responsible AI governance. The review also reveals that the principal challenge lies in the structural tension between agility and accountability, driven by the mismatch between technological and institutional change. Drawing on the Indonesian context, the study argues that governance agility should be operationalized through measurable indicators, institutional safeguards, and continuous evaluation, and proposes a future research agenda for conceptual refinement, empirical validation, comparative analysis, and governance measurement.]]>
Copyrights © 2026
