Definition
Vietnam's first implementing regulation under its AI Law, effective May 2026, which divides AI systems into high, medium, and low risk categories and attaches different compliance obligations to each. High-risk AI systems — those with significant potential for harm — face the strictest requirements.
Why it matters
Any company deploying AI in Vietnam — including multinationals — must classify their systems under this framework and meet category-specific obligations, making it a material compliance consideration for organisations with operations or customers in the country.