Country Profile on AI Governance: Somalia

Mean of the 8 scored indicators; 2 carry no score. East Africa mean 1.56 · all 54 states mean 1.76.

No score: 2 not scored (excluded from the mean).

1. S1.1

   National AI strategy existence

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Score without evidence; excluded from the pillar mean. Fellow scored 4 but the evidence reads "No verified evidence found"; every score must be supported by evidence.

2. S1.2

   Strategy adoption date

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean. Fellow entry "1900-01-04" is not a 0–4 score and no usable raw value was found.

3. S1.3

   Political commitment signals

   Score 1 / 4

   No official, published national AI strategy in force was identified. Somalia has a government-affiliated Somali National AI Center (SNAIC), founded in October 2024, and national AI convening has accelerated through the AI Summit Somalia 2026.

   Note: Score reflects emerging institutional readiness rather than an adopted national AI strategy.

   Sources snaic.gov.so (opens in a new tab)

4. S1.4

   Dedicated AI budget allocation

   Score 0 / 4

   No adoption date is available because no published/adopted national AI strategy was found. A related National Multi-Stakeholder Consultation Workshop validated Somalia’s Digital Transformation Strategy (2025–2030) on 12 October 2025, but this is not an AI-strategy adoption date.

   Note: Use 12 Oct 2025 only as evidence of digital-strategy consultation, not AI strategy adoption.

   Sources nca.gov.so (opens in a new tab)

5. S1.5

   AU Continental AI Strategy alignment

   Score 3 / 4

   Senior commitment is evidenced by the Second AI Summit Somalia 2026 convened by NCA, opened by the Minister of Education, with participation from the NCA Director General and State Minister of Communications and Technology; MoCT also participated in regional AI governance forums.

   Note: Strong political signals, but no presidential AI directive or adopted AI strategy located.

   Sources sonna.so (opens in a new tab)

6. S1.6

   Multi-stakeholder consultation

   Score 0 / 4

   No credible published source identified a ring-fenced national AI budget. Current evidence points to institution-building, consultations, and AI events rather than a published budget line for AI governance or development.

   Note: Do not infer an AI budget from broader digital transformation activity.

   Sources snaic.gov.so (opens in a new tab)

7. S1.7

   Sector prioritisation

   Score 2 / 4

   Somalia’s Digital Transformation Strategy consultation explicitly referenced regional alignment with African Union and East African Community frameworks. The AU Continental AI Strategy calls for member states to develop AI strategies/policies guided by ethical, responsible and equitable AI.

   Note: Alignment cannot be rated full because no Somalia AI strategy text is published for direct pillar-by-pillar mapping.

   Sources nca.gov.so (opens in a new tab)

8. S1.8

   Implementation roadmap

   Score 3 / 4

   NCA’s Digital Transformation Strategy validation brought together government ministries/agencies and licensed telecom operators. The AI Summit Somalia 2026 brought together government officials, universities, researchers, technology experts, innovators, entrepreneurs, civil society organizations and students.

   Note: Formal multi-stakeholder engagement exists, but the AI strategy-development process itself is not yet fully documented publicly.

   Sources nca.gov.so (opens in a new tab)

9. S1.9

   Monitoring and evaluation framework

   Score 3 / 4

   SNAIC identifies responsible AI priorities in healthcare, agriculture, governance and education, and lists focus areas including Somali NLP, agriculture, healthcare, education, livestock and blue economy. Somalia’s digital roadmap also targets innovation and inclusion across public services and economic sectors.

   Note: Sector prioritisation is inferred from SNAIC and digital-strategy sources, not from an adopted AI strategy document.

   Sources snaic.gov.so (opens in a new tab)

10. S1.10

    Strategy review mechanism

    Score 2 / 4

    The Digital Transformation Strategy consultation covered Somalia’s digital vision, strategic pillars, implementation roadmap, governance structures, funding models and regional alignment. No AI-specific phased implementation plan with milestones was found.

    Note: Score reflects a related digital roadmap, not a standalone AI roadmap.

    Sources nca.gov.so (opens in a new tab)

Fellow-assigned pillar score. Indicators not scored. East Africa mean 1.28 · all 54 states mean 1.32.

No score: 10 not scored (excluded from the mean).

1. G2.1

   AI-specific legislation

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

2. G2.2

   Data protection law status

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

3. G2.3

   DPA AI mandate

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

4. G2.4

   DPA staffing and budget

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

5. G2.5

   AI regulatory body existence

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

6. G2.6

   Institutional design model

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

7. G2.7

   Malabo Convention ratification

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

8. G2.8

   Enforcement actions taken

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

9. G2.9

   Extraterritorial provisions

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

10. G2.10

    IP framework for AI

    ○Not scored: Excluded from the pillar mean

    No verified evidence found

    Scoring basis Not determined; excluded from the pillar mean.

Mean of the 9 scored indicators; 1 carries no score. East Africa mean 1.54 · all 54 states mean 1.51.

No score: 1 context only.

1. I3.1

   Data center count

   Score 3 / 4

   At least 13 data centers are publicly evidenced/claimed: Hormuud operates 11 data centers; Dalkom provides colocation and web hosting from a Mogadishu data center; Eurosom advertises cloud servers hosted in a Mogadishu data center. Public data-center registries remain incomplete, with DataCenterMap listing only 1 Somalia facility, so this is a conservative evidence-based estimate rather than a complete national inventory.

   Note: Score reflects existing but still emerging/poorly documented data-center market. Hormuud count may include telco/internal sites, while Dalkom/Eurosom are clearer commercial cloud/colocation evidence.

   Scoring basis Threshold applied. Raw value: 13 (data centres).

   Sources datacenterdynamics.com (opens in a new tab)

2. I3.2

   Data center capacity (MW)

   Score 2 / 4

   The only clear published capacity figure found is Hormuud’s combined 10 MW data-center portfolio. Capacity for Dalkom, Eurosom and the National Data Center was not publicly disclosed in the reviewed sources.

   Note: Use 10 MW as a minimum documented capacity, not a full national total.

   Scoring basis Threshold applied. Raw value read from the evidence: "Hormuud’s combined 10 MW data-center portfolio" [single operator; fellow: "minimum documented capacity, not a full national total"]. Checked against the evidence on 7 October 2026: confirmed.

   Sources datacenterdynamics.com (opens in a new tab)

3. I3.3

   Submarine cable landing points

   Score 3 / 4

   Somalia has three confirmed submarine cable landing stations/points: Mogadishu, Berbera and Bosaso. NCA reported four cables (EASSy, Gulf2Africa, DARE and PEACE) terminating at these three landing stations, and 2025 regulations now govern submarine cables and landing-station licensees.

   Note: Strong connectivity asset for AI/data infrastructure, though resilience depends on domestic backbone and fair access regulation.

   Scoring basis Threshold applied. 5 cable systems (TeleGeography, accessed 2026-10-06).

   Sources nca.gov.so (opens in a new tab)

4. I3.4

   International bandwidth (Gbps)

   Score 2 / 4

   A recent World Bank Horn of Africa broadband analysis projected Somalia’s total international internet traffic at 635 Gbps in 2025 and 1,314 Gbps by 2030, but this is a projection rather than a current official measured total. The World Bank Eastern Africa Regional Digital Integration Project uses Somalia’s used international bandwidth baseline of 22.19, with a target of 40.00, expressed as used bandwidth per population rather than total national capacity.

   Note: Treat as Unknown/Estimated. No current NCA total Gbps figure was identified in accessible published sources.

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

   Sources documents1.worldbank.org (opens in a new tab)

5. I3.5

   Hyperscaler presence

   Score 0 / 4

   No AWS, Microsoft Azure, Google Cloud or other major global hyperscaler region/local zone was identified in Somalia. Official hyperscaler infrastructure pages list global footprints but no Somalia-based region or local zone. Local cloud services exist through Somali providers, but these are not global hyperscaler infrastructure.

   Note: Score is for global hyperscaler local infrastructure only, not general cloud service availability over the internet.

   Sources aws.amazon.com (opens in a new tab)

6. I3.6

   GPU/HPC access

   Score 1 / 4

   Somalia has emerging AI research institutions and labs, including SNAIC and Jamhuriya University’s Somali-language AI and Innovation Lab. However, no published evidence was found of a national GPU/HPC facility or commercially available local GPU/HPC cloud for AI research.

   Note: Limited academic/research AI capacity is present, but compute infrastructure remains weak or undocumented.

   Sources snaic.gov.so (opens in a new tab)

7. I3.7

   Data localisation provisions

   Score 2 / 4

   Somalia’s Data Protection Act No. 005 of 2023 establishes a general personal-data protection framework and applies to controllers operating in Somalia or processing/targeting Somali data subjects. Separately, the National Data Center is intended to store essential government data domestically and reduce reliance on foreign hosting. However, no blanket economy-wide data-localisation law was identified.

   Note: Score reflects general data-protection framework plus government-data localisation objective, not comprehensive localisation across all sectors.

   Sources dataguidance.com (opens in a new tab)

8. I3.8

   Cross-border data flow framework

   Score 4 / 4

   The Data Protection Act contains Chapter VI on cross-border transfers of personal data. It requires adequate protection or alternative legal bases/safeguards for transfers outside Somalia or to international organisations, and empowers the Data Protection Authority to issue guidance, adequacy designations and determinations. Somalia’s DPA is also operational as the national independent authority.

   Note: This is one of Somalia’s strongest P3 indicators because the legal transfer framework is explicit and institutionally anchored.

   Sources dataguidance.com (opens in a new tab)

9. I3.9

   Foreign AI surveillance spending

   ○Context only: Context flag outside the score

   Unknown. No credible public source was found documenting Somalia’s spending on foreign AI surveillance systems. A Ministry of Finance RFB exists for CCTV and access controls for NIRA HQs/enrollment centers, but the available public notice does not show AI features, foreign supplier details, or confirmed AI-surveillance spending. Recent Africa-wide reporting on AI-enabled surveillance spending identifies other African countries but did not identify Somalia in the accessible text.

   Note: Do not infer AI surveillance expenditure from ordinary CCTV/access-control procurement.

   Scoring basis Context flag; outside the score.

   Sources mof.gov.so (opens in a new tab)

10. I3.10

    Sovereign cloud initiatives

    Score 3 / 4

    Somalia’s National Data Center is under development in Mogadishu. NCA announced construction in 2023, and MoCT reported in May 2025 that construction was nearing completion; the center is intended to host government data and reduce dependence on foreign hosting, supporting data sovereignty.

    Note: Under development rather than operational, because no official operational launch was found in accessible sources.

    Sources nca.gov.so (opens in a new tab)

Mean of all 10 indicator scores. East Africa mean 1.18 · all 54 states mean 1.20.

1. H4.1

   AI researchers per million

   Score 1 / 4

   Unknown / not officially quantified. Evidence points to an emerging but small AI research base: SNAIC was founded in Oct 2024 as a governmental AI research/education institution; SIMAD University AI Institute and university AI labs/departments are active, but no public national researcher count exists.

   Note: Use conservative score. Numeric value should be recorded as Unknown until a national researcher registry or bibliometric count is available.

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

   Sources snaic.gov.so (opens in a new tab)

2. H4.2

   DPA staff with AI competency

   Score 0 / 4

   Unknown / not publicly documented. Somalia’s DPA was inaugurated in Feb 2024 and is operational, but no published staff list or evidence of AI-technical staff was found.

   Note: Score reflects lack of published evidence, not proof of absence.

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

   Sources dpa.gov.so (opens in a new tab)

3. H4.3

   University AI programs

   Score 2 / 4

   At least 3 documented AI/ML/Data Science degree-program pathways or dedicated departments: Jamhuriya University M.Sc. in AI & Data Science; Somali National University M.Sc. in Data Science with ML, Deep Learning and NLP courses; Hormuud University has launched an Artificial Intelligence department. Additional ML/data science offerings appear in some universities, but not all are fully verifiable as AI/ML degrees.

   Note: Minimum documented count. Excludes one-off short courses and general CS programs with only an ML course.

   Scoring basis Threshold applied. Raw value: 3 (universities).

   Sources just.edu.so (opens in a new tab)

4. H4.4

   African-authored AI publications

   Score 1 / 4

   Proxy estimate: AIWorld reports Somalia published 14 AI research papers in 2024. This is treated as a proxy because a more authoritative national bibliometric dataset specifying domestic institutional affiliation was not found.

   Note: Record numeric value as 14 (2024 proxy). Needs validation from OpenAlex/Scopus/Web of Science if available.

   Scoring basis Threshold applied. 14.0 papers/yr ÷ 19.01m people = 0.74 per million (floors applied).

   Sources aiworld.eu (opens in a new tab)

5. H4.5

   AI patents filed

   Score 0 / 4

   No AI-related domestic patent filings were identified. WIPO’s 2024 Somalia IP profile shows extremely limited patent activity, and the U.S. International Trade Administration notes Somalia lacks effective IP protection law.

   Note: Record as 0/Unknown for AI-specific patents. Conservative score is 0 due to no published AI-patent evidence.

   Scoring basis Threshold applied. Raw value read from the evidence: "No AI-related domestic patent filings were identified." [notes: "Record as 0/Unknown"]. Checked against the evidence on 7 October 2026: confirmed.

   Sources wipo.int (opens in a new tab)

6. H4.6

   K-12 CS education

   Score 3 / 4

   ICT has been introduced into the national curriculum at upper-primary level under the new Somali education system; however, this is broader ICT rather than a comprehensive computer science/AI curriculum and implementation capacity remains uneven.

   Note: Scored as mandatory in some grades, with caution that ICT is not the same as a full CS curriculum.

   Sources somalia.un.org (opens in a new tab)

7. H4.7

   AI workforce training programs

   Score 2 / 4

   Government-affiliated SNAIC lists AI education/certification programs, short-term AI workshops, certificate/diploma programs, university partnerships, and specialized courses for public officials and professionals. These appear emerging/limited rather than nationwide.

   Note: Score reflects operational/emerging capacity, not national-scale mandatory training.

   Sources snaic.gov.so (opens in a new tab)

8. H4.8

   Labour transition policy

   Score 1 / 4

   No adopted AI-specific labour transition framework was found. Somalia has employment policy work and UNESCO’s AI Readiness Assessment process is at inception, expected to identify human-capacity and policy reform priorities.

   Note: Scored as under discussion due to RAM/national AI policy-readiness work, not because a labour-transition policy exists.

   Sources unesco.org (opens in a new tab)

9. H4.9

   Informal economy AI assessment

   Score 0 / 4

   No government assessment of AI’s impact on Somalia’s informal economy was found. Published labour/informal economy studies document informality and skills gaps, but they do not assess AI-specific impacts or recommendations for informal workers.

   Note: Score 0. Useful context: Somalia’s labour market has high informality, so this is a priority evidence gap.

   Sources unescwa.org (opens in a new tab)

10. H4.10

    African language NLP coverage

    Score 4 / 4

    1 domestic African language with meaningful NLP/LLM support: Somali. Evidence includes SomBERTa for Somali fake-news/toxicity classification, SomaliWeb v1 corpus/tokenizer/benchmark, and Somali NLP focus by SIMAD/Jamhuriya AI labs.

    Note: Record numeric value as 1 language: Somali. Arabic is official in Somalia but excluded here because the indicator targets African language NLP coverage.

    Scoring basis Threshold applied. Raw value: 1 of 1 (languages supported / major languages), normalised 100. 100% of major domestic languages.

    Sources arxiv.org (opens in a new tab)

Mean of all 10 indicator scores. East Africa mean 0.97 · all 54 states mean 1.05.

1. N5.1

   AI companies count

   Score 2 / 4

   Approx. 30 AI startups/companies are reported by Tracxn’s Somalia AI startup database (Jan 2026). This is treated as a proxy because official company-registry AI classifications were not found.

   Note: Use with caution: third-party startup database; official Somalia AI company registry not found.

   Scoring basis Threshold applied. Raw value: 30 (companies).

   Sources tracxn.com (opens in a new tab)

2. N5.2

   Private AI investment (USD)

   Score 0 / 4

   Unknown. No credible published annual private AI investment figure for Somalia was found. General startup-funding proxies show a small ecosystem: StartupBlink reports Somalia startup funding over US$1.2m, and Innovate Ventures reports US$200k+ funding for Somali startups, but these are not AI-specific annual figures.

   Note: Record as Unknown for AI-specific investment; do not substitute general startup funding as AI investment.

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

   Sources startupblink.com (opens in a new tab)

3. N5.3

   AI incubators/accelerators

   Score 2 / 4

   At least two AI-focused or AI-explicit ecosystem support mechanisms are documented: SNAIC’s Startup Incubation & Acceleration programme and SIMAD AI Institute’s Innovation Hub/AI & Robotics Lab. Broader tech accelerators such as SIMAD iLab and Innovate Ventures support Somali startups but are not exclusively AI-focused.

   Note: Count is conservative: 2 AI-focused mechanisms; broader tech incubators noted separately.

   Scoring basis Threshold applied. Raw value: 2 (incubators/accelerators).

   Sources snaic.gov.so (opens in a new tab)

4. N5.4

   Regulatory sandbox (AI)

   Score 0 / 4

   No published AI-specific regulatory sandbox was identified. Somalia’s UNESCO AI Readiness Assessment is at inception and may inform future governance reforms, but it is not a sandbox.

   Note: Status: None found as of June 2026.

   Sources unesco.org (opens in a new tab)

5. N5.5

   Government AI procurement

   Score 1 / 4

   Early/ad hoc government-linked AI adoption support exists through SNAIC’s Public Sector AI Adoption Program, including technical support for ministries and AI-powered systems for healthcare, agriculture, education and civil services. However, no formal AI procurement framework or domestic-developer preference policy was found.

   Note: Classified as ad hoc/early-stage, not formal procurement.

   Sources snaic.gov.so (opens in a new tab)

6. N5.6

   AI safety evaluation capacity

   Score 1 / 4

   Academic and government-affiliated research capacity exists, including SNAIC’s ethical AI mission, SIMAD AI Institute’s AI governance/ethics work, and Jamhuriya University’s Somali-language AI Lab testing NLP/fake-news detection. No accredited AI safety testing lab or national evaluation facility was found.

   Note: Score reflects academic/research capacity only.

   Sources snaic.gov.so (opens in a new tab)

7. N5.7

   ISO/IEC 42001 adoption

   Score 1 / 4

   No Somalia-specific ISO/IEC 42001 adoption, certification, or pilot implementation was found. Somalia has a national standards body (SoBS) and participates in international standardization, while ISO/IEC 42001 is the global AI management-system standard.

   Note: Classified as aware/early standards capacity; adoption evidence not found.

   Sources iso.org (opens in a new tab)

8. N5.8

   Standards body AI competency

   Score 1 / 4

   The Somali Bureau of Standards exists and develops standards, with committees in areas such as food/agriculture, energy, construction and health, but no AI-specific technical committee was identified.

   Note: Score reflects existing standards body without documented AI technical committee.

   Sources sobs.gov.so (opens in a new tab)

9. N5.9

   AI risk assessment frameworks

   Score 1 / 4

   No sector-specific mandatory AI risk assessment framework was found. Somalia’s UNESCO AI Readiness Assessment is intended to identify institutional, regulatory and technical gaps and support ethical AI governance recommendations.

   Note: Classified as early/voluntary readiness work, not a framework in use.

   Sources unesco.org (opens in a new tab)

10. N5.10

    University-industry AI partnerships

    Score 3 / 4

    Joint labs/programs and formal collaboration signals exist: Jamhuriya University, in partnership with the University of Djibouti, launched the Somali-language AI and Innovation Lab (SAIL); SIMAD AI Institute operates a university-based AI hub with labs and innovation resources; SNAIC states partnerships with universities, private-sector technology companies and startups for joint ventures, incubators and accelerators.

    Note: Strongest evidence is joint labs/programs; industry component is emerging through SNAIC and innovation hubs.

    Sources dawan.africa (opens in a new tab)

Fellow-assigned pillar score. Indicators not scored. East Africa mean 1.16 · all 54 states mean 1.26.

No score: 10 not scored (excluded from the mean).

1. E6.1

   Algorithmic impact assessment

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

2. E6.2

   Bias audit requirements

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

3. E6.3

   Transparency obligations

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

4. E6.4

   Citizen redress mechanism

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

5. E6.5

   Judicial AI remedy

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

6. E6.6

   AI advisory body diversity

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

7. E6.7

   Global AI governance participation

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

8. E6.8

   Civil society AI engagement

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

9. E6.9

   African ethical framework reference

   ○Not scored: Excluded from the pillar mean

   No verified evidence found

   Scoring basis Not determined; excluded from the pillar mean.

10. E6.10

    Youth participation

    ○Not scored: Excluded from the pillar mean

    No verified evidence found

    Scoring basis Not determined; excluded from the pillar mean.

Mean of all 10 indicator scores. East Africa mean 1.54 · all 54 states mean 1.55.

1. R7.1

   AU AI Strategy alignment

   Score 2 / 4

   Somalia does not yet have a published national AI strategy, but the 2025–2030 Digital Transformation Strategy consultation explicitly included regional alignment with African Union and East African Community frameworks. UNESCO RAM implementation is also being used to produce a policy-oriented AI governance report.

   Note: Partial/formative alignment; no adopted AI-specific strategy alignment instrument found.

   Sources nca.gov.so (opens in a new tab)

2. R7.2

   Kigali Declaration endorsement

   Score 3 / 4

   Somalia is listed among the signatories of the Africa Declaration on Artificial Intelligence, adopted in Kigali on 4 April 2025. The declaration commits signatories to responsible national AI policy and governance aligned with the AU Continental AI Strategy.

   Note: Evidence is direct: Somalia appears in the declaration signatory list.

   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) · c4ir.rw (2) (opens in a new tab)

3. R7.3

   REC AI governance coordination

   Score 3 / 4

   Somalia participated in the UNESCO Eastern Africa Regional Forum on AI in Nairobi in June 2024. EAC regional AI coordination has advanced through EASTECO/IUCEA, including a Regional AI Network and Digital Leaders Forum focused on a draft EAC Regional AI Strategy.

   Note: Active participation, but no evidence that Somalia leads the REC process.

   Sources moct.gov.so (opens in a new tab)

4. R7.4

   Cross-border AI agreements

   Score 3 / 4

   Somalia and Saudi Arabia agreed in September 2025 to develop a cooperation framework on regulation of AI and space technology, including knowledge exchange, regulatory frameworks and infrastructure sharing.

   Note: Bilateral framework reported by Somalia's official news agency; implementation details not yet public.

   Sources sonna.so (opens in a new tab)

5. R7.5

   Election integrity AI provisions

   Score 1 / 4

   No AI-specific election integrity law or electoral AI provision was found. However, the Banadir local election process included media monitoring, a code of conduct, and journalist training with NIEBC/NUSOJ on election coverage and countering disinformation.

   Note: Institutional response exists for misinformation; AI-specific election safeguards remain a gap.

   Sources nusoj.org (opens in a new tab)

6. R7.6

   Disinformation response framework

   Score 1 / 4

   Somalia has ad hoc/general responses to harmful online content and election disinformation, including NIEBC/NUSOJ monitoring and earlier government action targeting TikTok and Telegram over harmful content. No government AI-generated disinformation framework was found.

   Note: Classified as ad hoc/general rather than AI-specific.

   Sources nusoj.org (opens in a new tab)

7. R7.7

   AI election incidents documented

   Score 2 / 4

   At least one AI-related election-period incident was documented: SJS reported an AI-generated audio clip carrying a false call for violence that was amplified during election campaign speeches in Mogadishu in December 2025.

   Note: Conservative count based on direct Somalia election-period reporting; broader misinformation levels were reported separately by NUSOJ.

   Scoring basis Threshold applied. Scored on documented institutional response, not incident count. Raw value read from the evidence: "At least one AI-related election-period incident was documented" [incidents: 1 (AI audio, Dec 2025); R7.5 NIEBC/NUSOJ media monitoring and code of conduct]. Listed for second-edition review: Somalia held no direct national election between 1 July 2021 and 30 June 2026. The incident relates to the Banadir local council elections of 25 December 2025, so the indicator may be not applicable.

   Sources sjsyndicate.org (opens in a new tab) · riotimesonline.com (opens in a new tab)

8. R7.8

   Platform content moderation

   Score 2 / 4

   Somali journalists' groups publicly called on Meta to identify, remove and limit AI-generated misinformation in Somalia. Government has also taken ad hoc action on social platforms over harmful content, but no formal AI-specific platform co-regulation mechanism was found.

   Note: Public pressure and ad hoc platform/content actions; no published structured AI moderation agreement.

   Sources sjsyndicate.org (opens in a new tab)

9. R7.9

   Geneva Dialogue participation

   Score 1 / 4

   The UN Global Dialogue on AI Governance is open to all governments and stakeholders, with the first session scheduled for 6–7 July 2026 in Geneva. No Somalia-specific participation or delegation announcement was found as of 18 June 2026.

   Note: Score reflects general eligibility/open invitation, not confirmed Somalia participation.

   Sources un.org (opens in a new tab)

10. R7.10

    G20/multilateral AI engagement

    Score 3 / 4

    Somalia is engaged in several multilateral/regional AI processes: it participated in the Eastern Africa AI forum, is listed as a signatory to the Africa Declaration on AI, and is undergoing UNESCO AI Readiness Assessment work to inform ethical AI governance.

    Note: Not a G20/BRICS participant; scored for AU/EAC/UNESCO multilateral AI engagement.

    Sources moct.gov.so (opens in a new tab)

Mean of the 9 scored indicators; 1 carries no score. East Africa mean 0.94 · all 54 states mean 0.78.

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

1. M8.1

   AI governance budget disbursed

   Score 0 / 4

   No published figure was found for actual AI-governance budget disbursed. Somalia has established SNAIC and is undertaking a UNESCO AI Readiness Assessment, but no public AI governance appropriation/disbursement report was identified.

   Note: Enter as 0/Unknown for budget value; score kept conservative because committed/disbursed AI-governance funds are not publicly reported.

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

   Sources snaic.gov.so (opens in a new tab)

2. M8.2

   Health AI regulation

   Score 1 / 4

   Somalia regulates health products and technologies through the Ministry of Health/NMRA framework, including medical devices, but no AI-specific health regulation or AI medical-device guidance was found. Health-AI oversight is therefore best treated as under discussion/emerging.

   Note: Base health-product regulation exists, but not yet an AI-specific health governance framework.

   Sources nmra.gov.so (opens in a new tab)

3. M8.3

   Health AI devices approved

   Score 0 / 4

   No public NMRA or Ministry of Health list/registry of formally approved AI health devices or AI clinical decision-support tools was found.

   Note: AI-enabled health pilots may exist, but formal national regulatory approval evidence was not found.

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

   Sources nmra.gov.so (opens in a new tab)

4. M8.4

   Government AI system registry

   Score 0 / 4

   The Somalia e-Government Strategy 2025-2029 and National Data Center work focus on digital public services, data centralisation, data sovereignty and public-service automation, but no public government AI-system registry was identified.

   Note: No evidence of a public inventory/registry of AI systems used by ministries, departments or agencies.

   Sources moct.gov.so (opens in a new tab)

5. M8.5

   AI governance M&E reports

   Score 0 / 4

   No published AI governance implementation/M&E report was found. UNESCO RAM work is intended to produce a policy-oriented country report and evidence base, but this is preparatory rather than a published implementation M&E report.

   Note: Use 0 until a national AI strategy/RAM report or implementation progress report is published.

   Scoring basis Threshold applied. Raw value: 0 (reports). Raw value read from the evidence: "No published AI governance implementation/M&E report was found." Checked against the evidence on 7 October 2026: confirmed.

   Sources unesco.org (opens in a new tab)

6. M8.6

   Agriculture AI governance

   Score 1 / 4

   Somalia’s agriculture policy direction includes digital agriculture, CIAD innovation/agripreneurship and digital agriculture ecosystem components, but no AI-specific agricultural governance provisions, standards or law were found.

   Note: Score reflects digital-agriculture governance foundations, not a completed AI-agriculture framework.

   Sources moa.gov.so (opens in a new tab)

7. M8.7

   Financial services AI governance

   Score 0 / 4

   The Central Bank of Somalia is modernising digital governance and financial-sector supervision, and financial-institution law provides supervisory powers; however, no AI-specific CBS guidance, sector AI regulation or enforcement action was found.

   Note: Financial digital-governance reforms are relevant but are not AI-specific.

   Sources centralbank.gov.so (opens in a new tab)

8. M8.8

   Education AI governance

   Score 1 / 4

   Somalia’s education sector planning includes ICT integration, digital learning and teacher digital-literacy initiatives, but no AI-specific education governance policy or guidelines were found.

   Note: Score reflects digital-learning foundations and emerging AI policy discussions, not adopted AI-in-education guidance.

   Sources education-profiles.org (opens in a new tab)

9. M8.9

   Measurable governance outcomes

   Score 1 / 4

   Somalia’s Data Protection Authority provides protective data-governance mechanisms relevant to future AI systems, including registration/licensing, complaints, breach notification, guidance and 2026 regulations; however, no AI-specific protective outcome or case was found.

   Note: Documented data-protection implementation exists, but measurable AI-governance outcomes remain early/anecdotal.

   Sources dpa.gov.so (opens in a new tab)

10. M8.10

    WHO AI health guidance adoption

    ○Not scored: Excluded from the pillar mean

    No verified evidence found

    Scoring basis Not determined; excluded from the pillar mean.