Dynamic Determinants of AI Readiness in ASEAN

A System GMM Approach on Productivity and Network Infrastructure

Abstract

Using a panel regression framework based on Difference and System GMM estimators developed by Arellano-Bond and Blundell-Bond, we investigated the dynamic relationship between AI Readiness, Productivity, and Network Infrastructure among six ASEAN member countries (Singapore, Malaysia, the Philippines, Indonesia, Thailand, and Vietnam). The findings of this study highlight that productivity growth remains the most critical driver of AI readiness in ASEAN, whereas enhancements in network infrastructure alone are insufficient without corresponding advancements in innovation capacity and digital skills. In addressing endogeneity and unobserved heterogeneity, the model conveys that lagged productivity exerts a strong and positive influence on AI readiness, while the effects of network infrastructure and AI implementation are mixed and, in some cases, statistically insignificant. Diagnostic tests, including the Arellano-Bond autocorrelation and Conditional Likelihood Ratio (CLR) test, confirm the validity and robustness of the instruments, confirming model reliability. The study contributes to the understanding of AI-driven development dynamics in emerging economies and supports policy formulation aligned with UN SDG 9 (Industry, Innovation, and Infrastructure) to advance sustainable technological growth and regional competitiveness.