AI Governance Capacity in Public Administration

This study compares artificial intelligence (AI) governance in public services in the European Union (EU) and Türkiye, focusing on institutional design and administrative capacity. Through a comparative institutional–legal analysis grounded in administrative capacity theory and a four-dimensional analytical framework, it examines the EU AI Act and Türkiye's sectoral governance arrangements to assess accountability, transparency, and enforceability in practice. The findings show that the EU has established a horizontal, risk-based governance framework, whereas Türkiye's approach remains characterized by sectoral fragmentation and a lack of a unified regulatory framework, with limited coordination and weak oversight capacity. The analysis identifies three main gaps: the absence of a formal risk classification system, unclear accountability arrangements, and insufficient institutional capacity for effective supervision. The study offers theoretically grounded recommendations to strengthen Türkiye's public-sector AI governance capacity in line with international standards.