Artificial Intelligence in Public Procurement in Developing Countries: A Narrative Policy Review and Implications for Vietnam

This article provides a policy review of the application of artificial intelligence (AI) in public procurement in developing countries, using Vietnam as an illustrative context. It is designed as a narrative policy review that combines the analysis of academic literature, reports from international organizations, and current legal documents. Using the public procurement cycle as the analytical framework, the article synthesizes AI applications in procurement planning, price forecasting, supplier risk assessment, bid document processing, bid-rigging detection, and contract implementation monitoring. The synthesis shows that AI can contribute to operational efficiency, transparency, and data-driven decision-making. However, the realization of these benefits depends substantially on data quality, model explainability, algorithmic audit mechanisms, data security, and the digital capacity of the public sector. The article proposes a four-pillar policy framework consisting of data foundations, algorithm governance, legal accountability, and operational capacity. For Vietnam, the adoption of AI in public procurement should follow a cautious, risk-based trajectory, beginning with low-risk support tasks before expanding to risk warning systems and decision-support applications with high legal implications.