Every AI Policy Ever · open in the interactive register
Published (non-binding, current)Non-binding · Guidelines / framework
FAIR Guidelines for the Development and Use of Artificial Intelligence
Record MU-01
- Issued by
- Ministry of Information Technology, Communication and Innovation (Artificial Intelligence Unit); adopted by Cabinet
- Jurisdiction
- Mauritius
- Level · region
- National · Sub-Saharan Africa
- Published / adopted
- 2026-04-09
- Language
- English
Summary
Non-binding national guidelines built on four pillars (Fairness, Accountability, Inclusiveness and Integrity, Responsibility) that accompany the National AI Strategy. They apply across sectors and the whole AI lifecycle, including systems from external providers, and set a risk-based governance model with indicative low, medium, high and unacceptable-risk tiers, escalation controls and safeguards, data-protection alignment, transparency, explainability and human oversight, accountability, auditability and record-keeping, procurement and vendor governance, and guidance on generative, foundation and agentic AI. They are to inform sector rules and future legislation where risk justifies it.
Safety topics
risk-based classificationhuman oversightaccountability & liabilitygeneral-purpose / frontier AItransparency & labelling
Status notes
Cabinet agreed on 13 February 2026 to adopt the national AI strategy, AI for Mauritius, as the national guiding policy framework, supported by the FAIR guideline document, and stated that the guidelines will evolve into AI regulations as the ecosystem matures. The strategy and the guidelines were launched on 9 April 2026 at the J & J Auditorium in Phoenix (the ministry's release of 10 April 2026 says 'launched yesterday'). The guidelines state that they are not a law and create no legal obligations, and they do not apply to AI used for military, defence, national security or counter-terrorism purposes.
Local copy
Archived. Last attempt 2026-10-18 04:08:03 UTC
Official source (third-party copy, not published) archive/MU-01/2026-10-11/official.pdf
Retrieved 2026-10-11 19:46:47 UTC
SHA-256 f025c3f55ca3b56fd46ee2015779ddb38e1c87c9af0941d30662cb1176986572
Secondary source (third-party copy, not published) archive/MU-01/2026-10-11/secondary.html
Retrieved 2026-10-11 19:47:04 UTC
SHA-256 e83595c90c716a33ec345945ba2ac97a6181060b08c18c02f88068c872371814
Verification
BAnother official source, such as an announcement by the issuer, confirms the title, date and status. checked 2026-10-18 · alternative-official-url-fetched
Recheck of 18 Oct 2026. Stored: the guidelines PDF from the AI Unit's site aim.govmu.org (38 pages; stored 11 Oct 2026 and unchanged on 18 Oct 2026), the ministry's release of 10 April 2026 on mitci.govmu.org, and the Highlights of the Cabinet meeting of 13 February 2026 from the Prime Minister's Office (pmo.govmu.org). The PDF confirms the title, the ministry, the non-binding character, the scope, the risk tiers, the scope exclusion and the legal trajectory; the release confirms the launch 'yesterday' (9 April 2026) and that the ministry's AI Unit led the work; the Cabinet highlights (item 6) confirm the adoption of the strategy 'AI for Mauritius' supported by the FAIR guideline document and that the guidelines will evolve into AI regulations. The PDF carries no date, so date_published rests on the ministry's release. pmo.govmu.org is not yet on the list of official sites, so the Cabinet quotes count as secondary until a person records the host. The same PDF is also hosted at mitci.govmu.org as 'FINAL-GUIDELINES-15AVRIL.pdf'.
Second check: Independent check by Claude Code (claude-sonnet-5-5), 2026-10-18: 7 field(s) supported.
History
- 2026-10-18verification.level: A → B — level computed from the evidence by tools/pipeline.py2026-10-18-recheck-mena-africa-1 · Claude Code (claude-opus-5-5)
- 2026-10-18notes: Cabinet agreed on 13 February 2026 to adopt the national AI strategy 'AI for Mau → Cabinet agreed on 13 February 2026 to adopt the national AI strategy, AI for Mau — kept the Cabinet adoption and its statement that the guidelines will evolve into AI regulations, now quoted from the stored Cabinet highlights of 13 February 2026 (item 6), and the launch date; added the scope exclusion for military, defence, national security and counter-terrorism uses (page 8); dropped the UNDP date, which no stored source carries; the remarks on the second hosted copy and on related strategies not in the register moved to verification.note and leads2026-10-18-recheck-mena-africa-1 · Claude Code (claude-opus-5-5)
- 2026-10-18summary: Non-binding national guidelines (Fairness, Accountability, Inclusiveness and Int → Non-binding national guidelines built on four pillars (Fairness, Accountability, — rewritten from the guidelines' own text: the closing sentence now gives the guidelines' own legal trajectory (pages 8 and 36) instead of the Cabinet statement, which stays in notes with its source; 'AI for Mauritius' replaced by the National AI Strategy as the guidelines name it2026-10-18-recheck-mena-africa-1 · Claude Code (claude-opus-5-5)
Citation
Ministry of Information Technology, Communication and Innovation (Artificial Intelligence Unit); adopted by Cabinet. FAIR Guidelines for the Development and Use of Artificial Intelligence. 2026-04-09. https://aim.govmu.org/aim/wp-content/uploads/2026/04/fairguidelines.pdf. Status: Published (non-binding, current); verified 2026-10-18 (level B). Local copy archive/MU-01/2026-10-11/official.pdf retrieved 2026-10-11 19:46:47 UTC, SHA-256 f025c3f55ca3b56fd46ee2015779ddb38e1c87c9af0941d30662cb1176986572. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record MU-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.