Conformity assessment mechanisms for high-risk AI systems under the EU AI Act and implications for Vietnam.

Abstract

In the context of Vietnam’s gradual development of a legal framework for the digital economy, managing the potential risks arising from artificial intelligence has become an urgent requirement. This article examines the experience of the European Union through the EU AI Act, with a particular focus on conformity assessment mechanisms for high- risk AI systems. By analyzing the management principles and regulatory processes of the European Union, the article identifies gaps in Vietnam’s current legal framework and, on that basis, proposes directions for developing control and quality assessment mechanisms for AI systems that are appropriate to the domestic context.