Explaining the Principle-Practice Gap: Governance Capability Asymmetry in Nepal's Emerging AI Governance Framework

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Background: Despite growing convergence around ethical AI principles, a persistent gap remains between high-level commitments and their operationalization, particularly in developing countries. Objectives: To examine the principle-practice gap as a governance capability challenge in Nepal’s emerging AI governance framework. Methods: A qualitative single-case study analyzed national policies, government decisions, sectoral frameworks, and supporting documentary evidence available from 2024 to mid-2026. Deductive thematic analysis used five New Public Governance (NPG) strands to identify governance arrangements and infer seven governance capability dimensions from their documentary specification. Results: Nepal has developed a broad AI governance architecture spanning all five NPG strands. Across seven governance capability dimensions, the documentary record shows greater specification of design capability than of mechanisms required to implement, enforce, steer, monitor, coordinate, and adapt governance arrangements depicting asymmetry in design and operationalization capability. Conclusion: In emerging AI governance framework characterized by technical complexity and distributed authority, the capability to design governance arrangements develops more extensively than the capabilities required to operationalize. Factors associated with it such as soft-law persistence, institutional immaturity, external dependence, limited smart-buyer capacity, coordination deficits, reliance on external expertise, and uneven institutional participation provide plausible governance-based explanations for the persistence of the principle-practice gap. Strengthening operational capability and matching governance arrangements to governance problems may help narrow the principle-practice gap. Novelty: The study integrates NPG and governance capability perspectives to examine how governance arrangements in an emerging AI governance framework reflect capability asymmetry and help explain the principle-practice gap.