Every AI Policy Ever · open in the interactive register
In forceBinding · Regulation / implementing rule
Nghị định số 142/2026/NĐ-CP quy định chi tiết một số điều và biện pháp thi hành Luật Trí tuệ nhân tạo
Record VN-04 · Decree No. 142/2026/NĐ-CP
Main implementing decree of Viet Nam's Law on Artificial Intelligence, detailing a number of its articles, from the one-stop AI portal and national database of AI systems to risk classification, conformity assessment and management of high-risk systems, controlled testing and impact assessment of AI used by state bodies. Providers classify systems before use; deployers reclassify with the provider when changes create new or higher risk. Providers must mark AI-generated audio, images and video in machine-readable form, and deployers must label content that imitates real people or events. Serious incidents must be reported within 72 hours or 5 working days, depending on the case.
No English version of this instrument was found. The English title and the summary are the register's own, written from the sources below; for the exact wording use the Vietnamese text.
Dated Hà Nội, 30 April 2026; the Government portal's digital signature on the published copy is dated 7 May 2026. Article 45 sets entry into force on 1 May 2026. Article 5(6) lets test, inspection and conformity results already obtained under sectoral, standards or product-quality law count toward the AI Law's requirements. The Ministry of Science and Technology manages and operates the portal (Article 4) and provides on it an optional automated risk-classification support tool (Article 10). Article 12 sets risk-classification dossiers for high- and medium-risk systems; Article 18 requires notice and visible labels for AI-generated or AI-edited content that could be mistaken for authentic; Article 19 defines serious incidents and sets preliminary reports within 72 hours for urgent cases and 5 working days for others, with a full report within 15 days; Chapter IV (Articles 21-27) sets a voluntary controlled-testing regime with three levels. Article 4(3)(d) exempts AI systems and databases serving the internal work of Party bodies and political organisations, and systems on the state-secret, defence and security lists, from notification, registration and public data sharing on the portal. The high-risk list is set separately by Decision 33/2026/QĐ-TTg.
Open the official source https://datafiles.chinhphu.vn/cpp/files/vbpq/2026/4/142-ndcp.signed.pdf
Secondary source: https://moc.gov.vn/vn/tin-tuc/1305/93854/nguyen-tac-phan-loai-va-danh-gia-su-phu-hop-he-thong-tri-tue-nhan-tao.aspx
Archived. The secondary source could not be downloaded (The site's TLS certificate could not be verified. robots.txt could not be checked, which RFC 9309 treats as 'do not crawl'). Last attempt 2026-10-14 10:15:13 UTC
Viet Nam Law on Artificial Intelligence (2025) Vietnam · In force
Viet Nam high-risk AI systems list (Decision 33/2026/QĐ-TTg) Vietnam · In force
Viet Nam AI Law implementation plan (Decision 367/QĐ-TTg) Vietnam · In force
Government of Viet Nam (drafted by the Ministry of Science and Technology). Decree No. 142/2026/NĐ-CP detailing a number of articles of, and measures for implementing, the Law on Artificial Intelligence [Decree No. 142/2026/NĐ-CP]. 2026-04-30. https://datafiles.chinhphu.vn/cpp/files/vbpq/2026/4/142-ndcp.signed.pdf. Status: In force; verified 2026-10-14 (level A). Local copy archive/VN-04/2026-10-11/official.pdf retrieved 2026-10-11 19:39:04 UTC, SHA-256 18b99b9bfdd4c9f237f84dfd16480735773f390b64381491404003745cdd5b6e. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record VN-04.
Know of an amendment, a status change or an error? Report news or an error on GitHub; the form opens with this record filled in.