Artificial intelligence and digitalization reinforce inequality in developing countries
Digitalization and artificial intelligence (AI) can expand access, productivity, and service delivery in developing countries, yet their benefits are distributed unevenly and may reinforce pre-existing structural inequalities. This systematic review distinguishes foundational digitalization from more computationally intensive AI applications and re-audits the evidence base in response to concerns about scope and reproducibility. Of the 75 publications retained in the broader literature corpus, 57 meet the prespecified criteria for the systematic synthesis; 18 are retained only as contextual/background literature and are not included in evidence counts. The review uses PRISMA 2020 reporting principles and a design-specific Mixed Methods Appraisal Tool (MMAT) procedure for eligible empirical studies. The thematic synthesis presents four interconnected mechanisms: unequal digital and AI capabilities, workforce and value-distribution pressures, uneven socio-economic results, and governance constraints. These are integrated into the ‘Vicious Cycle of Techno-Dependency’ as an interpretive framework, not a tested causal model. Evidence indicates that the strength and form of these mechanisms vary across country contexts and institutional capacities. Recent post-search studies from Brazil, India, and Indonesia further suggest that data-protection and sovereignty initiatives can strengthen accountability, but their effectiveness is conditional and remains incomplete. Policy priorities therefore include locally appropriate infrastructure and skills investment, participatory data governance, algorithmic accountability, and evaluation of implementation outcomes rather than reliance on formal policy adoption alone.
Authors
- Romny Ly
- Sokhom Ma
- Bora Ly
Institutions
- Cambodian Mekong University (KH)
- Paññāsāstra University of Cambodia (KH)
- ACLEDA University of Business (KH)
- Mekong University (VN)
Publication Details
- Journal
- Discover Global Society
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1007/s44282-026-00678-y
- Primary Topic
- Ethics and Social Impacts of AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00