The adoption of Artificial Intelligence in SMES of developing countries: A narrative literature review of barriers, enablers and economic impacts from a comparative perspective

Résumé

Abstract

Artificial intelligence is emerging as a strategic lever of primary importance for small and medium-sized enterprises (SMEs), particularly in developing countries where these entities constitute a pillar of emerging economies. Despite their central role, the ability of SMEs to adopt AI remains constrained by multidimensional factors (Pérez-Campdesuñer et al., 2025). Behavioral and structural barriers appear predominant in this context (Hammani & Ben Ali, 2026), while management commitment and employee adaptability are identified as key determinants of adoption (Lada et al., 2023; Tamanine et al., 2024).

This article proposes a narrative literature review based on thirty-eight studies published between 2021 and 2026, selected for their theoretical and empirical relevance. The analysis pursues four objectives: (1) conceptualize the notions of artificial intelligence, SMEs, and technological adoption in the context of developing countries; (2) analyze the theoretical foundations mobilized in the literature (TOE, TAM, UTAUT, RBV, Dynamic Capabilities, Schumpeterian theory); (3) synthesize empirical results to map the barriers, levers, and impacts of adoption; (4) compare adoption dynamics between developed and developing economies, while formulating practical recommendations for policymakers, SME leaders, and technology providers.

The results reveal that management commitment constitutes the most robust determinant of adoption, confirmed in various geographical contexts (Malaysia, Morocco, Saudi Arabia). Behavioral barriers—risk aversion, status quo bias—are particularly pronounced in emerging economies, contrasting with more regulatory and infrastructural barriers in advanced economies. The documented economic impacts (productivity gains, cost reduction, improved supply chain visibility) highlight the transformative potential of AI for SMEs, subject to the implementation of appropriate support policies.

Keywords :   Artificial intelligence, SMEs, technology adoption, developing countries, TOE, behavioral barriers, literature review.