VietnamAI regulation

Regulation Status

• Vietnam enacted its first comprehensive Law on Artificial Intelligence (Law No. 134/2025/QH15), passed by the National Assembly on 10 December 2025 (90.7% in favour) and in force from 1 March 2026 — the first such law in Southeast Asia; it sets a risk-based framework classifying AI systems by risk level, with transparency, labelling, testing, monitoring and human-oversight obligations on high-risk applications, prohibitions on harmful uses, and innovation mechanisms including a National AI Development Fund, AI vouchers and sandboxes, with oversight centralised under the Government and the Ministry of Science and Technology as lead coordinator.
• Vietnam now has a binding, comprehensive AI-specific law in force, complemented by the Digital Technology Industry Law and a growing body of implementing decrees and sectoral guidance.
• Implementation is well underway: the Prime Minister adopted an implementation plan via Decision No. 367/QĐ-TTg (February 2026) tasking the Ministry of Science and Technology to coordinate, the Government issued Decree No. 142/2026/ND-CP (May 2026) requiring vendors to classify AI systems into high/medium/low risk before deployment, with conformity assessments for high-risk systems and AI incident-reporting timelines, and Decision No. 33/2026/QD-TTg (July 2026) designates 46 high-risk AI systems across education, healthcare, banking, litigation, transportation, and ethnicity and religion, effective 15 August 2026.
• Decision No. 33/2026/QD-TTg requires strict management, human oversight and operational control for listed high-risk systems; existing systems in healthcare, education and finance must comply by 1 September 2027, while systems in the remaining listed sectors must comply by 1 March 2027, and state authorities may suspend non-compliant systems or those posing severe threats.
• The Law on Digital Technology Industry, passed in June 2025 and in force from 1 January 2026, legalises digital assets and establishes incentives for semiconductors, AI and digital startups, requiring AI systems to remain under human oversight and categorising them as high-risk, high-impact or non-high-risk; draft AI standards and sectoral rules (banking, higher education) are also progressing.