Research on AI Adoption for Productivity Improvement in the Public Sector
TL;DR
Examining how artificial intelligence is being adopted across eight APO member economies and the conditions required to translate task-level efficiency gains into broader public sector productivity improvements highlights a central paradox.
Abstract
AI Adoption for Productivity Improvement in the Public Sector examines how artificial intelligence is being adopted across eight APO member economies and the conditions required to translate task-level efficiency gains into broader public sector productivity. Drawing on evidence from Bangladesh, the Republic of China, India, Japan, the Republic of Korea, Malaysia, the Philippines, and Thailand, the publication explores AI deployment across government functions while examining the institutional factors that shape its impact. The study highlights a central paradox: although proven AI applications can deliver substantial efficiency gains in targeted processes, these gains do not automatically translate into system-wide productivity improvements. Governance, data integration, institutional coordination, accountability, skills, and resilience are critical to realizing AI’s broader value. The comparative analysis identifies four strategic positionings and offers practical insights for policymakers seeking AI adoption that is efficient, trusted, inclusive, and sustainable.