Abstract
Abstract
The 2030 Agenda for Sustainable Development calls for translating science and innovation into action rather than remaining at the theoretical level. However, evidence on whether scientific research translates into measurable sustainability outcomes remains limited, particularly for AI and digitalization research and in developing regions such as the Middle East and North Africa (MENA). This study examines whether scientific output in ‘AI and digitalization for sustainability’ is associated with improvements in the Sustainable Development Index (SDI) across 22 MENA countries. We employ a two-stage approach, a bibliometric analysis of publications indexed in Scopus and Web of Science over the period 2010–2025, followed by a dynamic panel model estimated using difference GMM linking these publications to SDI. The bibliometric results reveal a rapidly expanding but unevenly distributed research landscape, concentrated in a few leading countries and organized around three complementary themes: predictive climate analytics, AI and digitalization applications tied to energy and sustainability, and the managerial frameworks guiding their implementation. Econometrically, publications are positively associated with SDI performance, with the effect emerging only after a three-year lag, consistent with a knowledge-diffusion process. Governance plays a decisive moderating role, substantially amplifying this effect in well-governed countries while it becomes negligible in weakly governed ones. The study provides new empirical evidence on how AI and digitalization research contributes to sustainability outcomes in a rapidly digitalizing yet climate-vulnerable developing region.