Serhiy Lyeonov

Silesian University of Technology, Zabrze, Poland; 

Sumy State University, Sumy, Ukraine

serhiy.lyeonov@polsl.pl 

ORCID 0000-0001-5639-3008

Does government AI readiness translate into technological upgrading? Within–between evidence from manufacturing and exports

Roman Hruška

Faculty of Economics and Entrepreneurship,

Pan-European University, Bratislava, Slovakia

roman.hruska@paneurouni.com

ORCID 0009-0001-9712-512X

Yong Zhou

Sumy State University, Ukraine

ORCID 0009-0007-4479-979X

István Vajna

John von Neumann University, Doctoral School of Management and Business Administration, Hungary

steven.vajna@gmail.com

OCDID 0009-0006-9235-1060

Abstract. Artificial intelligence is increasingly viewed as a driver of productivity and technological transformation, yet it remains unclear whether national AI readiness translates into measurable upgrading of manufacturing and export structures. The study aims to examine how government AI readiness is associated with the technological structure of production and exports, distinguishing persistent between-country differences from short-run within-country changes and assessing prospective associations with technology-intensive manufacturing. The analysis combines the Oxford Insights Government AI Readiness Index with World Bank indicators for up to 176 countries in 2020, 2021, 2023 and 2024, applying two-way fixed effects, a within–between decomposition, non-linearity and interaction tests, ANCOVA and change-score models, and multiple robustness procedures. The core fixed-effects models show no significant contemporaneous GAIRI association, with coefficients ranging from 0.001 for logged ICT exports to 0.079 for high-technology exports within manufactured exports. By contrast, between-country GAIRI is positively associated with manufacturing value added (β=0.180), high-technology exports (β=0.519 and 0.491), and ICT service exports (β=0.168). Manufacturing exhibits only suggestive curvature, with a turning point of 57.59 and a formal inverted-U test of p=0.108. Higher 2021 GAIRI is prospectively associated with the 2023 machinery share (β=0.505, p=0.0016; pBH=0.0047).

Received: September, 2025

1st Revision: August, 2026

Accepted: September, 2026

DOI: 10.14254/2071-8330.2026/19-3/13

JEL Classification: O33, O38, L60, F14

Keywords: Government AI Readiness Index, artificial intelligence readiness, technological upgrading, manufacturing structure, high-technology exports, ICT service exports, within–between analysis