Country Profile on AI Governance: Mauritania

Mean of the 9 scored indicators; 1 carries no score. North Africa mean 2.23 · all 54 states mean 1.76.

No score: 1 not scored (excluded from the mean).

1. S1.1

   National AI strategy existence

   Score 3 / 4

   Mauritania adopted its National Artificial Intelligence Strategy 2024-2029 in March 2024, led by the Ministry of Digital Transformation, Innovation and Modernization of Administration (MTNIMA). A revised Draft 3 (2025-2029) is circulating. The strategy is adopted/launched but full cabinet-level enactment with line-item budget is not yet documented, so it sits at Established rather than Implemented.

   Note: Original 2024 launch confirmed by Maghreb Insider (15 Mar 2024) and Carnegie Endowment, which places Mauritania in the 2024 cohort alongside Ethiopia, Libya, Nigeria, and Zambia.

   Sources oecd.ai (opens in a new tab)

2. S1.2

   Strategy adoption date

   ○Not scored: Excluded from the pillar mean

   National AI Strategy formally unveiled in March 2024 (Draft published April 2024 PDF). Draft 3 (2025-2029 version) followed in 2025.

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "2024-03" is not a 0–4 score and no usable raw value was found.

   Sources maghrebinsider.com (opens in a new tab)

3. S1.3

   Political commitment signals

   Score 3 / 4

   Strategy implementation is led directly by MTNIMA at ministerial level; Minister Moctar Yedaly is the former AU Commission head of information society and a principal architect of the Malabo Convention, giving the political commitment unusual technical depth. However, no presidential decree or cabinet resolution formally elevating the strategy has been documented, so this is a strong formal statement rather than full institutional action.

   Sources lawgratis.com (opens in a new tab)

4. S1.4

   Dedicated AI budget allocation

   Score 1 / 4

   No dedicated, ring-fenced AI budget line item has been publicly disclosed. AI strategy activities appear to be embedded within MTNIMA's general digital transformation budget. The Digital-Y project (EUR 4 million, German cooperation) funds related e-government digitisation (Ijraati) but this is broader digital transformation funding, not an AI-specific allocation.

   Sources leconomistemaghrebin.com (opens in a new tab)

5. S1.5

   AU Continental AI Strategy alignment

   Score 3 / 4

   The strategy explicitly aligns with AU Continental AI Strategy priorities: human capacity building, research and innovation, data governance, ethical AI, and sectoral applications (health, education, agriculture, transport, energy, defence). Mauritania also ratified the Malabo Convention and endorsed the Africa Declaration on AI (Kigali, April 2025), demonstrating sustained continental alignment beyond the strategy document itself.

   Sources oecd.ai (opens in a new tab)

6. S1.6

   Multi-stakeholder consultation

   Score 2 / 4

   No publicly documented formal multi-stakeholder consultation process (civil society, academia, private sector) comparable to processes seen in peer countries (e.g. Mauritius's UNDP-supported consultations) has been identified for Mauritania's AI strategy development. The strategy was developed primarily through MTNIMA with limited public evidence of broad-based consultation, though Draft 3 revisions suggest some iterative review.

   Note: Absence of evidence is treated as a Developing-level score rather than zero, since iterative drafting (Draft 1 to Draft 3) implies some internal review process.

   Sources dig.watch (opens in a new tab)

7. S1.7

   Sector prioritisation

   Score 3 / 4

   Strategy explicitly identifies priority sectors: healthcare, education, agriculture, fishing, transport, energy, and defence, with AI expected to support resource optimisation, service quality, and national security in each.

   Note: Strongest areas are political commitment, stakeholder engagement, and ethical governance.

   Sources oecd.ai (opens in a new tab)

8. S1.8

   Implementation roadmap

   Score 2 / 4

   Draft 3 (2025-2029) is organised around six core strategic priorities including human capacity building, research/innovation financing, and international cooperation, indicating a structured roadmap. However, no publicly available phased timeline with named milestones, owners, and dates has been published, so this remains a generic-to-structured roadmap rather than a fully accountable phased plan.

   Note: Areas requiring further evidence include monitoring, evaluation, and publication of implementation progress reports.

   Sources dig.watch (opens in a new tab)

9. S1.9

   Monitoring and evaluation framework

   Score 1 / 4

   No published Monitoring & Evaluation framework or reporting mechanism for AI strategy implementation has been identified. Strategy implementation is nominally overseen by a sub-committee under the Supreme Council for Digitalization, but no public M&E reports or progress disclosures exist as of the assessment date.

   Sources lawgratis.com (opens in a new tab)

10. S1.10

    Strategy review mechanism

    Score 2 / 4

    The existence of a Draft 3 (2025-2029) revision of the original 2024-2029 strategy demonstrates an active internal review and update process, though this appears ad hoc/ministerial-driven rather than a formally scheduled or legislated review mechanism.

    Sources dig.watch (opens in a new tab)

Mean of the 9 scored indicators; 1 carries no score. North Africa mean 1.56 · all 54 states mean 1.32.

No score: 1 not scored (excluded from the mean).

1. G2.1

   AI-specific legislation

   Score 1 / 4

   No dedicated AI-specific legislation exists in Mauritania. AI governance is currently addressed through the National AI Strategy (a policy document, not law) and the pre-existing Law No. 2017-020 on Personal Data Protection, which predates AI-specific concerns. Mauritania has also participated in CIS Model Law discussions on AI technologies as an observer-type engagement.

   Note: Mauritius has made significant progress through the National AI Strategy 2025–2029 and the FAIR Guidelines, but AI governance remains largely policy-driven rather than statute-driven.

   Sources lawgratis.com (opens in a new tab)

2. G2.2

   Data protection law status

   Score 2 / 4

   Law No. 2017-020 on Personal Data Protection was enacted in 2017 establishing foundational legal principles for personal data processing. However, it is hindered by the lack of implementing decrees, which constrains enforcement. The Commission Nationale de Protection des Donnees a Caractere Personnel (CNDCP/APD) exists but lacks full operational capacity, so this sits between Draft and fully Enforced.

   Note: The strongest governance institutions are the AI Unit and the Data Protection Office.

   Sources lawgratis.com (opens in a new tab)

3. G2.3

   DPA AI mandate

   Score 1 / 4

   The CNDCP/APD has no documented explicit AI oversight mandate. Its core mandate is general personal data protection under Law 2017-020; there is no public evidence of guidance, statements, or engagement specifically addressing AI systems.

   Note: There is currently no dedicated AI legislation, no AI-specific enforcement record, and no publicly available evidence of AI-related IP law reform.

   Sources lawgratis.com (opens in a new tab)

4. G2.4

   DPA staffing and budget

   ○Not scored: Excluded from the pillar mean

   No official government publication, annual report, or parliamentary disclosure provides CNDCP staffing or budget figures. Estimate is derived by peer-benchmarking against comparable West African LDC-context DPAs in early operational phases (typically 10-30 staff) and known patterns of Mauritanian regulatory body budget allocation. This is a reasoned estimate, not a verified figure, and is flagged Low Confidence.

   Note: Numeric entry field; treat figure as estimated/unverified pending CNDCP disclosure.

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "Not publicly disclosed (estimate: <15 FTE staff; <USD 500,000 annual budget)" is not a 0–4 score and no usable raw value was found.

5. G2.5

   AI regulatory body existence

   Score 2 / 4

   No standalone AI governance body exists. AI strategy implementation is managed through a sub-committee under the Supreme Council for Digitalization (Conseil Superieur du Numerique) -- a coordination mechanism under formation/operation, but without independent statutory authority or enforcement powers.

   Sources lawgratis.com (opens in a new tab)

6. G2.6

   Institutional design model

   Score 2 / 4

   The institutional model is a hybrid: the existing DPA (CNDCP/APD) handles general data protection while the Supreme Council for Digitalization sub-committee coordinates AI strategy implementation. No new dedicated statutory AI body has been created, and the division of authority between the two is not clearly codified in public documentation.

   Sources lawgratis.com (opens in a new tab)

7. G2.7

   Malabo Convention ratification

   Score 4 / 4

   Mauritania ratified the African Union Convention on Cyber Security and Personal Data Protection (Malabo Convention) on 9 May 2023, becoming the decisive 15th ratifying state and triggering the Convention's entry into force on 8 June 2023. Under Article 80 of Mauritania's Constitution, ratified treaties have supremacy over ordinary domestic law, giving the Convention direct legal force.

   Sources dataprotection.africa (opens in a new tab)

8. G2.8

   Enforcement actions taken

   Score 0 / 4

   No publicly documented AI-related enforcement action -- investigation, fine, or remedial order -- has been identified for Mauritania in any sector.

9. G2.9

   Extraterritorial provisions

   Score 1 / 4

   No comprehensive domestic legal framework addressing foreign AI systems or extraterritorial data processing has been identified beyond the general principles in Law 2017-020. Malabo Convention ratification creates a continental-level obligation but domestic implementing legislation specifically addressing extraterritorial AI system operation has not been enacted.

   Sources lawgratis.com (opens in a new tab)

10. G2.10

    IP framework for AI

    Score 0 / 4

    No evidence of IP legislation review or amendment specifically addressing AI-generated content has been identified in Mauritania.

Mean of the 8 scored indicators; 2 carry no score. North Africa mean 1.63 · all 54 states mean 1.51.

No score: 1 not scored (excluded from the mean), 1 context only.

1. I3.1

   Data center count

   Score 0 / 4

   No commercial data center of meaningful scale has been publicly documented in Mauritania. Digital infrastructure investment to date has focused on fibre backbone and connectivity (National Digital Transformation Agenda) rather than data center construction. No hyperscaler or colocation facility presence identified.

   Scoring basis Threshold applied. Raw value: 0 (data centres). Raw value read from the evidence: "No commercial data center of meaningful scale has been publicly documented in Mauritania." [absence of documented commercial DC]. Checked against the evidence on 7 October 2026: confirmed.

   Sources dig.watch (opens in a new tab)

2. I3.2

   Data center capacity (MW)

   Score 0 / 4

   No data center power capacity figures are publicly available, consistent with the absence of documented commercial data center infrastructure.

   Scoring basis Fellow score kept; no raw value for the threshold.

3. I3.3

   Submarine cable landing points

   Score 2 / 4

   No submarine cable landing point count specific to Mauritania was identified in available sources. Mauritania's international connectivity strategy, per the National Digital Transformation Agenda, includes plans to strengthen connectivity through submarine and terrestrial cables, implying current landing point capacity is limited, but an exact count could not be verified.

   Note: Flag for verification against ARTP (Autorite de Regulation) technical filings if available.

   Scoring basis Threshold applied. 2 cable systems (TeleGeography, accessed 2026-10-06).

   Sources dig.watch (opens in a new tab)

4. I3.4

   International bandwidth (Gbps)

   ○Not scored: Excluded from the pillar mean

   No verified figure for total international internet bandwidth (Gbps) was identified in available sources.

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "Not publicly disclosed" is not a 0–4 score and no usable raw value was found.

5. I3.5

   Hyperscaler presence

   Score 1 / 4

   No major global hyperscaler (AWS, Microsoft Azure, Google Cloud) local infrastructure presence has been identified in Mauritania. The country's three telecom operators (Chinguitel, Mattel, Moov Mauritel) provide core connectivity but not cloud hyperscale services locally.

   Sources techafricanews.com (opens in a new tab)

6. I3.6

   GPU/HPC access

   Score 1 / 4

   No national GPU cluster, high-performance computing facility, or dedicated AI compute infrastructure has been identified. Any AI model development activity (e.g. fintech algorithmic tools) likely relies on cloud-based compute accessed remotely rather than domestic HPC facilities.

7. I3.7

   Data localisation provisions

   Score 1 / 4

   No comprehensive data localisation law has been identified. Law No. 2017-020 on Personal Data Protection contains general principles but no explicit sectoral or comprehensive data localisation mandate has been documented.

   Sources lawgratis.com (opens in a new tab)

8. I3.8

   Cross-border data flow framework

   Score 2 / 4

   Mauritania's principal cross-border data framework derives from the Malabo Convention (ratified 2023), which establishes continental-level principles for data protection and cross-border cooperation. No additional bilateral data-transfer agreements specific to Mauritania were identified, so this sits at a regional-framework level rather than comprehensive bilateral coverage.

   Sources dataprotection.africa (opens in a new tab)

9. I3.9

   Foreign AI surveillance spending

   ○Context only: Context flag outside the score

   No documented or estimated figure for foreign AI surveillance system spending in Mauritania was identified in publicly available sources.

   Scoring basis Context flag; outside the score.

10. I3.10

    Sovereign cloud initiatives

    Score 0 / 4

    No national sovereign cloud or domestic computing initiative has been identified in Mauritania's digital transformation documentation. Infrastructure investment to date prioritises connectivity (fibre backbone expansion, satellite licensing tender) over sovereign compute capacity.

    Sources techafricanews.com (opens in a new tab)

Mean of the 9 scored indicators; 1 carries no score. North Africa mean 1.54 · all 54 states mean 1.20.

No score: 1 not scored (excluded from the mean).

1. H4.1

   AI researchers per million

   ○Not scored: Excluded from the pillar mean

   No official statistic for AI researchers per million population exists for Mauritania. Given the absence of dedicated AI degree programmes at Universite de Nouakchott Al-Aasriya and limited published AI research output, the figure is assessed as very low relative to regional leaders (Morocco, Tunisia).

   Note: Flag for verification; no national R&D survey data identified.

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "Not publicly disclosed (estimated low single digits per million)" is not a 0–4 score and no usable raw value was found.

2. H4.2

   DPA staff with AI competency

   Score 0 / 4

   No public evidence that any CNDCP/APD staff possess demonstrable AI technical competency; staffing levels themselves are not disclosed.

   Scoring basis Fellow score kept; no raw value for the threshold.

3. H4.3

   University AI programs

   Score 0 / 4

   Universite de Nouakchott Al-Aasriya is the primary academic institution; the National AI Strategy commits to 'enhance university AI capacities and launch new programs' (Objective 1.2), indicating this is a stated intention rather than an existing dedicated AI/ML degree programme.

   Scoring basis Threshold applied. Raw value: 0 (universities). Raw value read from the evidence: "a stated intention rather than an existing dedicated AI/ML degree programme". Checked against the evidence on 7 October 2026: confirmed.

   Sources slideshare.net (opens in a new tab)

4. H4.4

   African-authored AI publications

   Score 2 / 4

   No AI publications with Mauritanian-institution-affiliated authors were identified in available bibliographic or research-output searches.

   Scoring basis Threshold applied. 6.3 papers/yr ÷ 5.17m people = 1.23 per million (floors applied). Raw value read from the evidence: "No AI publications with Mauritanian-institution-affiliated authors were identified" [source used: "bibliographic or research-output searches"]. Checked against the evidence on 7 October 2026: confirmed.

5. H4.5

   AI patents filed

   Score 0 / 4

   No AI-related patent filings attributable to Mauritanian domestic entities have been identified.

   Scoring basis Threshold applied. Raw value: 0 (AI patent applications/year). Raw value read from the evidence: "No AI-related patent filings attributable to Mauritanian domestic entities have been identified." Checked against the evidence on 7 October 2026: confirmed.

6. H4.6

   K-12 CS education

   Score 1 / 4

   No evidence of a structured K-12 computer science curriculum (mandatory or elective) was identified for Mauritania. Digital literacy initiatives appear concentrated at the post-secondary and workforce level (per the AI Strategy's human capacity priorities) rather than primary/secondary education.

   Sources oecd.ai (opens in a new tab)

7. H4.7

   AI workforce training programs

   Score 2 / 4

   The National AI Strategy's first priority pillar is human capacity building, including training and certification programmes across academic and professional levels (per Draft 3, 2025-2029), and the Kosmos Innovation Center (established 2018 by mining companies Kinross, MCM, and SNIM) supports entrepreneurship and skills training, though not AI-specific at present.

   Sources dig.watch (opens in a new tab)

8. H4.8

   Labour transition policy

   Score 1 / 4

   No dedicated government framework for AI-driven labour market transition has been identified. The general unemployment context (Ministry of Employment and Youth estimates ~37% versus the ILO-modelled rate of 10.37% for 2024) indicates labour market policy is currently focused on basic employment generation rather than AI-specific transition planning.

   Sources state.gov (opens in a new tab)

9. H4.9

   Informal economy AI assessment

   Score 0 / 4

   No government assessment of AI's impact on Mauritania's large informal economy (agriculture, artisanal fishing/mining, animal husbandry) has been identified or published.

   Sources state.gov (opens in a new tab)

10. H4.10

    African language NLP coverage

    Score 0 / 4

    Hassaniya Arabic, Mauritania's primary spoken dialect, has no significant publicly documented corpus or NLP/LLM support at scale. Standard Arabic and French have some general-purpose LLM coverage via major international models, but Hassaniya-specific support is effectively absent, representing both a structural gap and an identified strategic opportunity.

    Scoring basis Threshold applied. Raw value: 0 of 4 (languages supported / major languages), normalised 0. 0% of major domestic languages. Raw value read from the evidence: "Hassaniya-specific support is effectively absent" [Pulaar, Soninke, Wolof not discussed; denominator (set by the editors, not taken from the profile text): Hass. Checked against the evidence on 7 October 2026: confirmed.

    Sources aviaanaccounting.com (opens in a new tab)

Mean of the 9 scored indicators; 1 carries no score. North Africa mean 1.39 · all 54 states mean 1.05.

No score: 1 not scored (excluded from the mean).

1. N5.1

   AI companies count

   Score 2 / 4

   No official AI company registry exists. African Business reports fewer than 100 tech-enabled startups active nationwide, concentrated in Nouakchott, spanning digital/IT services, fintech, e-commerce, education, and early-stage AI applications. The AI-specific subset is estimated at 5-20 entities based on the total startup population and the nascent AI skills base, with no documented AI-dedicated venture funding rounds identified.

   Note: Estimate flagged Low Confidence; no official registry exists.

   Scoring basis Threshold applied. Raw value: 12.5 (companies). Raw value read from the evidence: "The AI-specific subset is estimated at 5-20 entities" [midpoint of 5-20; fellow estimate; directory used: African Business startup total]. Listed for second-edition review: The raw value is the midpoint of a fellow estimate of 5 to 20 companies.

   Sources african.business (opens in a new tab)

2. N5.2

   Private AI investment (USD)

   ○Not scored: Excluded from the pillar mean

   No documented AI-specific private investment figures for Mauritania were identified. The broader startup ecosystem is characterised by bootstrapping; founders typically maintain parallel employment due to absence of reliable external financing.

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "Not publicly disclosed (likely negligible)" is not a 0–4 score and no usable raw value was found.

   Sources african.business (opens in a new tab)

3. N5.3

   AI incubators/accelerators

   Score 0 / 4

   No AI-focused incubator or accelerator program has been identified in Mauritania. The Kosmos Innovation Center (2018, mining-sector backed) supports general entrepreneurship and small business development but is not AI-specific.

   Scoring basis Threshold applied. Raw value: 0 (incubators/accelerators). Raw value read from the evidence: "No AI-focused incubator or accelerator program has been identified in Mauritania." Checked against the evidence on 7 October 2026: confirmed.

   Sources state.gov (opens in a new tab)

4. N5.4

   Regulatory sandbox (AI)

   Score 0 / 4

   No AI-specific regulatory sandbox has been identified or announced in Mauritania.

5. N5.5

   Government AI procurement

   Score 1 / 4

   No formal government AI procurement framework exists. Government e-government platforms (Khidmaty, Ijraati) appear to be developed via direct ministerial digitisation initiatives (MTNIMA, supported partly by German cooperation through the Digital-Y project) rather than a competitive domestic AI procurement process.

   Sources leconomistemaghrebin.com (opens in a new tab)

6. N5.6

   AI safety evaluation capacity

   Score 0 / 4

   No AI safety testing or evaluation facility, academic or governmental, has been identified in Mauritania.

7. N5.7

   ISO/IEC 42001 adoption

   Score 0 / 4

   No evidence of ISO/IEC 42001 (AI management systems standard) adoption, awareness campaigns, or pilot implementation has been identified for Mauritania.

8. N5.8

   Standards body AI competency

   Score 0 / 4

   No evidence that Mauritania's national standards body maintains an AI-specific technical committee or working group was identified.

9. N5.9

   AI risk assessment frameworks

   Score 0 / 4

   No sector-specific or cross-sector mandatory AI risk assessment framework has been identified in Mauritania.

10. N5.10

    University-industry AI partnerships

    Score 1 / 4

    No formal university-industry AI research partnership (MoU or joint lab) has been identified. The closest analogue is the long-standing mining-sector-backed School of Mining (collaborating with Ecole Polytechnique Montreal since 2011) and the Kosmos Innovation Center, neither of which is AI-specific.

    Sources state.gov (opens in a new tab)

Mean of all 10 indicator scores. North Africa mean 1.33 · all 54 states mean 1.26.

1. E6.1

   Algorithmic impact assessment

   Score 0 / 4

   No algorithmic impact assessment requirement, voluntary or mandatory, has been identified in Mauritania. The CNDCP/APD has no documented mandate for algorithmic impact assessment.

   Sources lawgratis.com (opens in a new tab)

2. E6.2

   Bias audit requirements

   Score 0 / 4

   No AI bias audit requirement for high-risk systems exists in Mauritania's legal or regulatory framework.

3. E6.3

   Transparency obligations

   Score 0 / 4

   No AI-specific transparency obligation has been identified in domestic law or regulation.

4. E6.4

   Citizen redress mechanism

   Score 1 / 4

   Citizens have access to a general data protection complaint channel in principle via the CNDCP/APD under Law 2017-020, but no AI-specific complaint mechanism exists, and the DPA's limited operational capacity (lack of implementing decrees, unclear staffing/budget) constrains the practical accessibility of this informal redress channel.

   Note: Government also operates Himayeti (human rights violation reporting) and Ain (administrative complaints/corruption reporting) platforms via Khidmaty, which could theoretically capture AI-adjacent grievances, but neither is AI-specific.

   Sources lawgratis.com (opens in a new tab)

5. E6.5

   Judicial AI remedy

   Score 0 / 4

   No documented judicial precedent or established court jurisdiction over AI-related disputes has been identified in Mauritania.

6. E6.6

   AI advisory body diversity

   Score 0 / 4

   No national AI advisory body with documented membership exists in Mauritania (AI strategy governance runs through a ministerial sub-committee rather than a multi-stakeholder advisory body), so diversity cannot be assessed.

   Sources lawgratis.com (opens in a new tab)

7. E6.7

   Global AI governance participation

   Score 2 / 4

   Mauritania participates in regional and international AI governance forums: it endorsed the Africa Declaration on Artificial Intelligence at the Kigali Global AI Summit (April 2025) and participated in the Turkic States AI Forum (Bishkek, October 2024) and the CIS Model Law on AI Technologies approval process. This constitutes active participation rather than mere observer status, but no leadership role (e.g. hosting, chairing) has been documented.

   Sources lawgratis.com (opens in a new tab)

8. E6.8

   Civil society AI engagement

   Score 0 / 4

   No formal mechanism for civil society participation in AI governance has been identified. Civil society organisations focused on rights issues (e.g. IRA-Mauritanie on anti-slavery advocacy) report facing registration and operating restrictions rather than being incorporated into AI policy consultation processes.

   Sources unpo.org (opens in a new tab)

9. E6.9

   African ethical framework reference

   Score 0 / 4

   No reference to African ethical frameworks (e.g. Ubuntu philosophy) or similar indigenous ethical principles has been identified in Mauritania's AI strategy or governance documentation.

10. E6.10

    Youth participation

    Score 0 / 4

    No formal mechanism for youth participation in AI governance has been identified in Mauritania.

Mean of all 10 indicator scores. North Africa mean 1.47 · all 54 states mean 1.55.

1. R7.1

   AU AI Strategy alignment

   Score 3 / 4

   Mauritania's National AI Strategy explicitly aligns with AU Continental AI Strategy priorities (human capacity, research/innovation, data governance, ethical AI, sectoral applications). This represents an active alignment process; full formal certification of complete alignment by the AU has not been documented.

   Sources oecd.ai (opens in a new tab)

2. R7.2

   Kigali Declaration endorsement

   Score 3 / 4

   Mauritania endorsed the Africa Declaration on Artificial Intelligence at the Global AI Summit on Africa in Kigali, Rwanda, on 4 April 2025, one of 49 African countries (per Carnegie Endowment's count) plus the African Union and Smart Africa as institutional co-signatories. No documented evidence yet of active implementation steps specifically tied to the Declaration (e.g. participation in the proposed USD 60 billion Africa AI Fund) beyond endorsement.

   Note: Note: widely circulated reporting states 'all 55 AU member states' endorsed; Carnegie's more precise figure of 49 countries is used here as the more rigorous source.

   Scoring basis Official signatory list applied. Signatory of the Africa Declaration on AI (Kigali, 4 April 2025), C4IR list: "Endorsed" = 3.

   Sources c4ir.rw (opens in a new tab) · carnegieendowment.org (opens in a new tab)

3. R7.3

   REC AI governance coordination

   Score 2 / 4

   Mauritania holds dual regional status: full membership in the Arab Maghreb Union (AMU/UMA) and observer status (not full membership) in ECOWAS, plus founding membership in the G5 Sahel security/development framework. No documented evidence of active AI-specific governance coordination at the REC level (AMU or ECOWAS) was identified; participation appears limited to general continental fora (AU, Smart Africa) rather than REC-specific AI working groups.

   Sources worldbank.org (opens in a new tab)

4. R7.4

   Cross-border AI agreements

   Score 0 / 4

   No bilateral or multilateral AI governance cooperation agreement specific to Mauritania has been identified, despite a clear strategic opportunity given the shared Greater Tortue Ahmeyim (GTA) gas project with Senegal and Minister Yedaly's continental network from his AU Commission background.

5. R7.5

   Election integrity AI provisions

   Score 1 / 4

   No specific legal or institutional provision addressing AI use in elections has been identified. The 2024 presidential election (won by incumbent President Ghazouani) was reportedly marred by some fraud and censorship allegations per civil society reporting, but this relates to general electoral integrity concerns rather than documented AI-specific interference or governance response.

   Sources unpo.org (opens in a new tab)

6. R7.6

   Disinformation response framework

   Score 0 / 4

   No government framework specifically addressing AI-generated disinformation has been identified in Mauritania.

7. R7.7

   AI election incidents documented

   Score 0 / 4

   No documented AI-related election interference incidents have been identified for Mauritania in available sources.

   Scoring basis Threshold applied. Scored on documented institutional response, not incident count. Raw value read from the evidence: "No government framework specifically addressing AI-generated disinformation has been identified in Mauritania." [incidents: 0; R7.5 none identified]. Checked against the evidence on 7 October 2026: confirmed.

8. R7.8

   Platform content moderation

   Score 0 / 4

   No formal or informal engagement between the Mauritanian government and major digital platforms specifically on AI content moderation for the domestic context has been identified.

9. R7.9

   Geneva Dialogue participation

   Score 1 / 4

   No confirmed delegation or participation plan for the UN Global Dialogue on AI Governance (July 2026) has been publicly announced for Mauritania as of the assessment date, though the country's active record of engagement with continental and international AI fora (Kigali Declaration, CIS Model Law process, Turkic States AI Forum) suggests at minimum an observer-level interest is plausible.

   Note: Score reflects inferred plausibility, not confirmed registration; flag for verification closer to the July 2026 Dialogue.

   Sources lawgratis.com (opens in a new tab)

10. R7.10

    G20/multilateral AI engagement

    Score 1 / 4

    Mauritania's most notable multilateral AI-adjacent engagement to date is its participation in the Turkic States Artificial Intelligence Forum (Bishkek, October 2024) and the CIS Model Law on AI Technologies approval process -- both observer/participant-level engagements rather than G20, BRICS, or comparable agenda-setting forums, in which Mauritania does not hold membership.

    Sources lawgratis.com (opens in a new tab)

Mean of the 9 scored indicators; 1 carries no score. North Africa mean 1.05 · all 54 states mean 0.78.

No score: 1 not scored (excluded from the mean).

1. M8.1

   AI governance budget disbursed

   ○Not scored: Excluded from the pillar mean

   No published figures distinguishing AI governance budget committed versus actually disbursed have been identified for Mauritania, consistent with the absence of a dedicated, ring-fenced AI budget line (see P1.S1.4).

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "Not publicly disclosed" is not a 0–4 score and no usable raw value was found.

2. M8.2

   Health AI regulation

   Score 0 / 4

   No AI-specific healthcare regulation has been identified in Mauritania. The National AI Strategy names healthcare as a priority sector for future AI application, but no implementing guidelines or legislation specific to health AI have been enacted.

   Sources oecd.ai (opens in a new tab)

3. M8.3

   Health AI devices approved

   Score 0 / 4

   No AI health device or diagnostic tool has been documented as formally approved by Mauritanian regulatory authorities.

   Scoring basis Threshold applied. Raw value: 0 (authorised devices). Raw value read from the evidence: "No AI health device or diagnostic tool has been documented as formally approved by Mauritanian regulatory authorities." Checked against the evidence on 7 October 2026: confirmed.

4. M8.4

   Government AI system registry

   Score 0 / 4

   No government registry of AI systems in use (public or private sector) has been identified for Mauritania.

5. M8.5

   AI governance M&E reports

   Score 0 / 4

   No published Monitoring & Evaluation report on AI governance implementation has been identified for Mauritania.

   Scoring basis Threshold applied. Raw value: 0 (reports). Raw value read from the evidence: "No published Monitoring & Evaluation report on AI governance implementation has been identified for Mauritania." Checked against the evidence on 7 October 2026: confirmed.

6. M8.6

   Agriculture AI governance

   Score 1 / 4

   Agriculture is named as a priority sector in the National AI Strategy, and Mauritania's structural opportunity in Sahel pastoral/climate-risk AI is significant given its geographic position, but no specific AI governance provisions (guidelines or legislation) for the agricultural sector have been enacted to date -- this remains at the under-discussion/strategic-intent stage.

   Sources oecd.ai (opens in a new tab)

7. M8.7

   Financial services AI governance

   Score 1 / 4

   No central bank (Banque Centrale de Mauritanie) guidance specifically on AI in financial services has been identified. The fintech sector (Bankily, Masrvi, Sadad Mauritanie) operates under general banking and electronic payments regulation rather than AI-specific financial services governance.

   Sources thefintechtimes.com (opens in a new tab)

8. M8.8

   Education AI governance

   Score 0 / 4

   No AI-specific governance provision for the education sector has been identified, despite education being named a priority sector in the National AI Strategy.

   Sources oecd.ai (opens in a new tab)

9. M8.9

   Measurable governance outcomes

   Score 0 / 4

   No documented case or systematic evidence of AI governance producing measurable protective outcomes (e.g. prevented harms, enforcement-driven behaviour change) has been identified for Mauritania.

10. M8.10

    WHO AI health guidance adoption

    Score 0 / 4

    No formal adoption of the WHO Ethics and Governance of Artificial Intelligence for Health guidance has been documented for Mauritania.