Vietnam has enacted Law No. 134/2025/QH15 on Artificial Intelligence, introducing a comprehensive regulatory framework governing the development, deployment, and use of AI systems. The law aims to promote innovation while ensuring safety, security, and ethical use of AI technologies.

A central feature of the law is its risk-based approach, classifying AI systems according to their potential impact and imposing stricter obligations on high-risk applications. Key requirements include transparency, human oversight, risk management, and safeguards against bias and misuse. The law also reinforces data protection and cybersecurity considerations in AI deployment.

The regulation will become fully effective on March 1, 2026, marking the official operational start of AI governance in Vietnam. This provides a transition period for organizations to align their systems and processes with the new legal requirements.

Vietnam’s Law No. 134/2025/QH15 on Artificial Intelligence represents a high-impact regulatory development as it introduces a comprehensive, cross-sector legal framework governing the development, deployment, and use of AI systems. Effective March 1, 2026, the law establishes binding obligations that directly affect market access, compliance requirements, and operational practices across multiple industries.

• Risk-Based Regulatory Framework: Introduces classification of AI systems (high, medium, low risk), with stricter compliance obligations for high-risk applications, impacting design, testing, and deployment.
• New Compliance Obligations: Requires transparency, human oversight, risk management processes, and accountability mechanisms for AI developers and providers.
• Cross-Sector Application: Applies broadly across industries including healthcare, finance, telecom, and digital platforms, significantly expanding regulatory scope.
• Cybersecurity and Data Protection Integration: Embeds security and data governance requirements into AI lifecycle management, increasing compliance complexity.
• Ethical and Safety Standards: Mandates safeguards against bias, discrimination, and misuse, requiring redesign or adjustment of existing AI systems.
• Clear Enforcement Timeline: Comes fully into force on March 1, 2026, requiring organizations to rapidly align operations and compliance frameworks.

Who?

• AI Developers and Providers: Must redesign systems to meet risk classification, transparency, and governance requirements.
• Manufacturers of AI-Enabled Devices: Need to ensure embedded AI complies with new legal and safety standards.
• Technology Companies and Digital Platforms: Face increased compliance obligations, especially for high-risk AI applications.
• Healthcare, Finance, and Telecom Sectors: Heavily impacted due to use of high-risk AI systems subject to stricter controls.
• Importers and Distributors: Responsible for ensuring AI-enabled products meet regulatory requirements before market entry.
• Regulatory Authorities: Gain expanded oversight and enforcement responsibilities.
• End Users / Consumers: Benefit from enhanced protections, transparency, and safer AI systems.

Conclusion

Law No. 134/2025/QH15 is a high-impact regulatory milestone that positions Vietnam at the forefront of AI governance in the region. Its entry into force on March 1, 2026 marks a significant shift toward structured oversight of artificial intelligence, requiring substantial adaptation from industry stakeholders while strengthening trust, safety, and accountability in AI deployment.