Vietnam’s Law on Artificial Intelligence (AI), which takes effect on March 1, 2026, stipulates that service providers must inform users when they are interacting with AI; otherwise, they may be subject to criminal proceedings or administrative fines.
The Ministry of Science and Technology is drafting a decree to provide clearer rules on AI labeling. According to the draft, AI systems that interact directly with humans must inform users that they are engaging with AI; AI-generated content must carry appropriate labels; and content that simulates a real person’s appearance or voice, or recreates real events, must include clear and distinguishable labeling.
For artistic and creative works, labels must not disrupt the delivery of the content.
Exceptions include AI content for internal use, such as within B2B and B2E ecosystems without public access; experiences conducted in controlled environments; and imaginative works where audiences clearly understand that the content is fictional. Other exceptions apply to AI-enhanced content—such as improvements to images, sound, appearance, or text—that does not alter the intended message.
Regarding labeling requirements, AI-supported news must display a clear notice on the screen or beneath the headline to ensure visibility. For short videos, labels must appear in the corners of the screen or in captions provided by the platform. For audio content, labels must be delivered as a voice notice at the beginning. For images, any watermark and metadata, where applicable, must be included as part of the labeling.
Additionally, AI service providers and content users must not remove or interfere with labels. Social media platforms are responsible for supporting users in labeling content, but they are not required to proactively review user contents, except in cases of violations or upon formal requests from authorities.