Every AI Policy Ever · open in the interactive register
In forceBinding · Law
Mākslīgā intelekta centra likums
Record LV-01 · Latvijas Vēstnesis 2025/55.1 (19.03.2025)
Establishes the foundation Mākslīgā intelekta centrs (AI Centre) to build Latvia's AI ecosystem and reduce AI-related risks, so that AI is used ethically, responsibly and safely respecting fundamental rights. Its tasks include identifying AI security risks in the public and private sectors, raising risk-management literacy, drafting AI guidelines, proposing measures against risks to human rights, democratic governance or public safety and, from 1 June 2026, giving opinions on AI and deepfake use in election campaigns. It runs a regulatory sandbox with the competent authorities in which legal requirements may be set aside for a period, provided testing does not endanger safety or health.
English translation by Valsts valodas centrs (State Language Centre) Open it. The Latvian text is authoritative.
Adopted by the Saeima on 6 March 2025, promulgated on 19 March 2025 and in force from 20 March 2025, the day after promulgation. The centre had to be founded by 31 March 2025, the Cabinet's rules on the sandbox and on data processing in it were due by 31 May 2025, and the task of giving opinions on AI and deepfakes in elections applies from 1 June 2026. The council has nine members: one each from the Ministry of Smart Administration and Regional Development (VARAM), the Ministry of Economics and the Ministry of Defence, and members nominated by the private sector and the universities; the Saeima named the Latvian Information and Communication Technology Association and the Latvian Universities Association as the private and academic founders. VARAM supervises the centre's delegated public tasks. An amending law of 14 May 2026, in force from 1 June 2026, deleted Article 7(5). The law does not refer to Regulation (EU) 2024/1689; Latvia's AI Act authorities are set out in the Cabinet's 2025 implementation report (LV-02).
Open the official source https://likumi.lv/ta/id/359339
Secondary source: https://www.saeima.lv/en/news/saeima-news/34443-saeima-approves-the-creation-of-an-artificial-intelligence-technology-ecosystem-in-latvia
Archived. A source changed since the previous snapshot; both versions are kept. Last attempt 2026-10-17 21:36:48 UTC
Latvia AI Act implementation report (2025) Latvia · Published (non-binding, current)
Saeima of the Republic of Latvia. Artificial Intelligence Centre Law [Latvijas Vēstnesis 2025/55.1 (19.03.2025)]. 2025-03-06. https://likumi.lv/ta/id/359339. Status: In force; verified 2026-10-17 (level A). Local copy archive/LV-01/2026-10-12/official.html retrieved 2026-10-12 04:38:01 UTC, SHA-256 0651920f126607ab22245eb3a97764059be39181aab5f07a141880e854f15c0e. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record LV-01.
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.