Every AI Policy Ever · open in the interactive register
Published (non-binding, current)Non-binding · Strategy / action plan
National Artificial Intelligence (A.I.) Policy, 2025
राष्ट्रिय आर्टिफिसियल इन्टेलिजेन्स (ए.आई.) नीति, २०८२
Record NP-01
- Issued by
- Government of Nepal, Council of Ministers, on the proposal of the Ministry of Communication and Information Technology
- Jurisdiction
- Nepal
- Level · region
- National · South & Southeast Asia
- Published / adopted
- 2025
- Language
- Nepali
Summary
Nepal's national AI policy. It commits the government to a risk-management and AI governance framework in line with globally recognised standards, including a framework to classify AI systems by risk and mitigate the risks. Action policies include AI laws to assess risks and ensure responsible and accountable use; standards for data, algorithms and technologies; benchmarking and certification of AI systems for quality, safety and reliability; standards against cybersecurity risks and AI-driven misinformation; a regulatory sandbox; and measures against deepfakes and personal biases. An AI Regulation Council chaired by the ICT minister is to develop standards and regulate AI.
Language
English translation by Government of Nepal Open it. The Nepali text is authoritative.
Safety topics
risk-based classificationaccountability & liabilityevaluation & testingdeepfakes & synthetic mediapublic-sector useresearch priorities
Status notes
The Cabinet endorsed the policy, presented by the Ministry of Communication and Information Technology, at its meeting of 11 August 2025, and the government spokesperson announced the decision on 14 August 2025 (The Himalayan Times, 14 August 2025); the policy text itself is undated. The Nepal Law Commission lists the policy under 6 Asoj 2083 (22 September 2026). The AI Regulation Council is chaired by the Minister for Communication and Information Technology and comprises the secretaries of the Office of the Prime Minister and Council of Ministers and of the ICT, finance and education ministries, the chair of the Nepal Telecommunications Authority, the CEO of the E-Governance Board, three experts from the private sector and universities (at least one a woman) and the chief of the National AI Centre as member secretary; it meets at least twice a year, and the National AI Centre serves as its secretariat. The Ministry reviews implementation annually, and the policy is to be reviewed every two years. Section 14 on risk identification concerns risks to the policy's implementation; the classification of AI systems by risk is to come from an AI governance framework under action policy 9.53.
Local copy
Archived. Last attempt 2026-10-18 03:22:59 UTC
Official source (third-party copy, not published) archive/NP-01/2026-10-12/official.pdf
Retrieved 2026-10-12 04:44:12 UTC
SHA-256 c8dad816f35f455fe04973d648f3326d660dc9316943114b68a08203bce47eb2
Secondary source (third-party copy, not published) archive/NP-01/2026-10-18/secondary.html
Retrieved 2026-10-18 00:49:01 UTC
SHA-256 3bbe1a9cfb128e5ef405310481642cd0121f099035bb7782d493afb152fac42f
Verification
AThe instrument's own official text, or an official register, was stored and confirms the title, date and status. checked 2026-10-18 · official-url-fetched
Recheck of 18 Oct 2026. Stored: the English text of the policy, 'National Artificial Intelligence (A.I.) Policy, 2025' (56 pages including the Schedule-1 action plan), from the Government of Nepal's web platform giwmscdnone.gov.np (copy of 12 Oct 2026, re-checked unchanged on 18 Oct 2026); the Nepal Law Commission's page for the policy (lawcommission.gov.np, 18 Oct 2026), which gives the Nepali title and the listing date 6 Asoj 2083; the Nepali text embedded in that page (giwmscdnone.gov.np, 32 pages; its text layer is garbled by a legacy font encoding, so it is not quoted); remarks of Nepal's Foreign Secretary of 7 December 2025 on mofa.gov.np ('The Cabinet's approval of the National AI Policy 2082'); and The Himalayan Times of 14 August 2025 (secondary, saved with a browser). The English text confirms the title, vision, goals, policies 7.1-7.6, strategies 8.1-8.11, action policies 9.1-9.66, the AI Regulation Council, the National AI Centre, the AI Excellence Centres and sections 11-16; it names no translator and carries no date. The official text is the Nepali one, so language is now Nepali and the English text is recorded as a government translation. date_published is given at the precision of the stored official texts (the year 2025 in the English title); the Cabinet date of 11 August 2025, the Ministry's role in presenting the policy and the announcement of 14 August 2025 rest on The Himalayan Times. The record's official_url is the English translation, not the Nepali text.
Second check: Independent check by Claude Code (claude-sonnet-5-5), 2026-10-18: 9 field(s) supported.
History
- 2026-10-18english_version: – → {"kind": "government translation", "url": "https://giwmscdnone.gov.np/media/pdf_ — added: the English text on the Government of Nepal's web platform (the record's official_url), a government translation2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18notes: The Cabinet endorsed the policy at its meeting of 11 August 2025 (press reports; → The Cabinet endorsed the policy, presented by the Ministry of Communication and — the Council's composition completed (experts, member secretary) and the two-yearly review added from the text; the statement that AI-system risk classification is left to future laws and standards corrected (action policy 9.53 assigns it to an AI governance framework); the claims on 'an official English version' on the Law Commission page and on the press report of the National AI Centre's inauguration removed, since no stored file shows them; the source of the Cabinet date (The Himalayan Times) named2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18safety_topics: ["accountability & liability", "evaluation & testing", "deepfakes & synthetic me → ["risk-based classification", "accountability & liability", "evaluation & testin — risk-based classification added (action policy 9.53: an AI governance framework to classify AI systems by risk)2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18summary: Nepal's first national AI policy, endorsed by the Cabinet in August 2025. It com → Nepal's national AI policy. It commits the government to a risk-management and A — the unsupported 'first' and 'endorsed by the Cabinet in August 2025' removed; the risk-classification framework of action policy 9.53, the quality, safety and reliability aim of 9.5 and the measures against deepfakes and personal biases of 9.60 added from the text2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18language: Nepali, English → Nepali — the policy was adopted by the Cabinet of Nepal; the Nepal Law Commission lists it under its Nepali title and embeds a 32-page Nepali text, stored in this run; the English text names no translator, so it is a translation, not an official text2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18date_published: 2025-08-11 → 2025 — the stored official text gives only the year (2025, in its title) and the Law Commission page gives only its own listing date; the day 11 August 2025 is the Cabinet meeting reported by The Himalayan Times and moves to notes (AGENTS.md section 7.7: date at the precision of the official source)2026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
- 2026-10-18title: National Artificial Intelligence (A.I.) Policy, 2082 (2025) → National Artificial Intelligence (A.I.) Policy, 2025 — the English text's own title; the published title mixed the Nepali year 2082 with the English text's 2025 ('2082 (2025)'), which no source uses; the short title keeps 20822026-10-18-recheck-south-asia-1 · Claude Code (claude-opus-5-5)
Citation
Government of Nepal, Council of Ministers, on the proposal of the Ministry of Communication and Information Technology. National Artificial Intelligence (A.I.) Policy, 2025. 2025. https://giwmscdnone.gov.np/media/pdf_upload/National%20AI%20Policy-Final_uxc94vg.pdf. Status: Published (non-binding, current); verified 2026-10-18 (level A). Local copy archive/NP-01/2026-10-12/official.pdf retrieved 2026-10-12 04:44:12 UTC, SHA-256 c8dad816f35f455fe04973d648f3326d660dc9316943114b68a08203bce47eb2. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record NP-01.
Improve this record
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.