Vietnam Leads the Way: Southeast Asia’s First AI Law Enters Force

Vietnam has become the first country in Southeast Asia to move AI governance from aspiration to law.

The AI Law, adopted by the National Assembly in December 2025, came into force in March 2026 under a four-year phased implementation schedule. It takes a risk-based approach modelled broadly on the EU’s AI Act: systems are classified by the potential harm they could cause, with higher-risk applications facing stricter requirements around transparency, accountability, and human oversight.

Across Southeast Asia, AI governance has until now relied largely on voluntary frameworks and soft guidelines. Singapore has its AI Verify framework; Thailand and Indonesia have published draft principles. Vietnam is the first to put binding rules on the statute books.

The obligations are real. Any organisation deploying AI systems in Vietnam now faces requirements to disclose when content is AI-generated, conduct impact assessments for high-risk applications, and maintain records of system performance and decision-making. The four-year phased rollout means the most demanding requirements do not kick in immediately — but the clock has started.

Vietnam’s move will likely accelerate legislative timelines elsewhere. Indonesia is expected to issue presidential AI regulations in 2026. The Philippines has signalled it will propose a framework during its ASEAN chairmanship this year. Thailand’s AI legislation remains at the consultation stage. For multinationals with technology operations across the region, five or six different national AI laws — each with its own classification criteria, documentation requirements, and enforcement mechanisms — is where this is heading.

The harder question for Vietnam is enforcement. Drafting a law and implementing it are different things. The country still needs to appoint the regulatory authorities responsible for oversight, build technical inspection capacity, and publish guidance that helps businesses interpret their obligations. How substantive the law turns out to be will depend heavily on what happens in the next twelve months.

Vietnam’s AI Law, in force since March 2026, makes it Southeast Asia’s first jurisdiction with binding AI rules. Businesses deploying AI in Vietnam now face disclosure, impact assessment, and record-keeping requirements. With Indonesia, the Philippines, and Thailand all advancing their own frameworks, AI compliance in the region is about to get more complicated — and more fragmented.