Africa Focus · Nigeria
Nigeria’s National AI Strategy, explained
Africa’s largest economy has a five-year plan to become a global AI leader. Around it sits a binding regulatory environment that is already in force and is moving faster than the strategy itself. This explains both, and says plainly where the gaps are.
Status verified 21 August 2026The strategy was published September 2025. Every pending instrument below was re-checked against primary and legislative sources before publication, because a pending bill described as law is exactly the error this page exists to prevent.
5Strategic pillars
34Strategies to implement the vision
2025–2029Implementation horizon
70%Young workforce to be AI-skilled
The NAIS was published in September 2025 by the Federal Ministry of Communications, Innovation and Digital Economy, co-created by NCAIR and NITDA. It is framed inside the wider national ambition of a trillion-dollar economy by 2030, with AI positioned as a structural lever rather than an add-on.
It sets three goals — economic growth and competitiveness, social development and inclusion, and technological advancement and leadership — and rests on nine principles including responsible conduct, transparency, human-centricity, risk management and data protection.
Primary source:National Artificial Intelligence Strategy, 19 September 2025 (PDF, ncair.nitda.gov.ng)
The strategy
Five pillars, 34 strategies
What each pillar sets out to do, and an honest read on where it stands.
01Foundational AI infrastructure
Affordable high-performance computing at national scale, expanded data centres, AI clusters powered primarily by clean energy with under 50% grid dependence, and Pioneer status tax incentives to pull private capital into the compute gap.
Reality check Nigeria’s AI infrastructure index is 42.67. The clean-energy target is ambitious given grid instability — but explicitly designing around the power problem is more honest than strategies that ignore it. The absence of a hyperscaler region inside Nigeria remains the single biggest structural gap: AWS, Azure and Google Cloud route Nigerian workloads through South Africa or Europe.
A world-class AI ecosystem
70% of the 16–35 workforce AI-skilled, including 50% of women and people with disabilities. AI curricula from secondary school upward. A talent transfer programme with diaspora return pathways, deep tech accelerators, and sector AI Centres of Excellence.
Reality check NAIRS is already funding 45 startups and researchers, the EduAI Hub is live, NitHub is active. The foundation exists. But Nigeria trains significant AI talent and most of it emigrates — retention, not training, is the crux. Without genuine financial incentives in the Talent Transfer Programme, upskilling risks accelerating brain drain rather than solving it.
Adoption and sector transformation
Lighthouse projects in agriculture, healthcare, education and finance. An Open Data Initiative, a National AI R&D Fund, a research and education network, and Green AI as a formal objective with its own challenge and grant programme.
Reality check Sector breadth is ambitious, but the strategy wisely anchors early implementation on agriculture and healthcare — deepest development-challenge density, richest existing data, and NAIRS grants already flowing there. Green AI as a formal objective rather than a mention is ahead of comparable African strategies.
Responsible and ethical AI
A High-Level AI Ethics Expert Group and National AI Ethics Commission, a lifecycle ethics assessment framework, and a horizon scan on AI’s labour-market impact. Nigerian ethical principles led by decoloniality, humanity, inclusion and responsible data governance. NIST AI RMF adopted as the primary international reference.
Reality check The decoloniality principle is intellectually distinctive and positions Nigeria’s ethics framework deliberately apart from Western-centric approaches. But the Ethics Commission is proposed, not constituted. Until it is formally established and staffed with real technical and civil-society expertise, the ethics framework exists on paper.
A robust AI governance framework
National AI Principles, an independent AI Governance Regulatory Body promoted by NITDA, a National AI Policy Framework and a National AI Risk Management Framework — built on the AIGN Framework, a modular and certifiable governance system drawing on the EU AI Act and NIST RMF.
Reality check Well-structured and internationally grounded. The regulatory body’s independence is the critical test: NITDA has significant commercial and policy interests in AI, and a body it promotes but does not control is the right design — provided the governance of that independence holds.
What actually binds you
Nigeria’s AI regulatory stack
These instruments are independent of the strategy and form the enforcement environment around it. Compliance is mandatory now, regardless of how implementation of the strategy progresses.
NDPA 2023 — Section 37
In forceHuman oversight required for automated decisions with material effect on individuals. Data subjects can contest AI-driven outcomes.
NDPC GAID 2025
In force — Sep 2025Mandatory DPIAs for high-risk AI, annual impact assessments, 72-hour breach notification, DPO requirement for data controllers of major importance.
NITDA National Cloud Policy 2025
In force — Oct 2025Level 3/4 sovereign data — government, financial, health, security — must be hosted in Nigeria. NITDA approval required for cross-border transfers.
CBN AML/AI Directive
In force — Mar 2026Banks have an 18-month AI/AML deployment deadline. All institutions had 90 days to file an implementation roadmap with the CBN. Annual independent model validation required.
NCC Internet Code of Practice
In force — Feb 2026Notify the NCC before deploying AI in telecoms operations. 48-hour breach notification — stricter than the NDPA’s 72-hour window.
Digital Economy & E-Governance Bill
Not lawPassed the House of Representatives by May 2026 and awaits Senate consideration. If enacted it would make NITDA the AI risk classifier and introduce developer licensing and risk-based scrutiny. Penalty figures reported in secondary coverage are unverified against the primary bill text — do not plan against them.
The strategy does not reference the GAID, the CBN directive, the Cloud Policy or the NCC Code as components of itself. They were enacted independently. That separation is the practical point: the binding environment is ahead of the strategy, and it is what an enterprise is actually assessed against.
The strategy’s own risk register
What Nigeria says could go wrong
Ethical concerns carry the only Extreme rating in the register — higher probability and higher impact than any other risk the strategy names.
| Risk | Probability | Impact | Rating |
|---|
| Ethical concerns | High | High | Extreme |
| Funding shortfall | Medium | High | High |
| Lack of skilled workforce | Medium | High | High |
| Data security breaches | Medium | High | High |
| Rapid technological change | High | Medium | High |
| Slow private-sector adoption | Medium | Medium | Medium |
The most distinctive risk the strategy names is digital colonialism — foreign models trained on non-Nigerian data dominating local deployment and embedding foreign values and biases. It sits under the Extreme-rated ethical concerns, and it is the reason the decoloniality principle exists.
Independent assessment
What it gets right, and what is missing
An analytical read, not an official position. Drawn from research conducted for a board-level AI governance baseline across Sub-Saharan Africa.
Gets right
- A final published strategy with a real process behind it — 120+ stakeholders, eight working groups, 18 months from workshop to publication.
- Ethical concerns carry the register’s only Extreme rating. Putting ethics at the top of the risk register is a meaningful signal, not decoration.
- Ethical principles anchored in decoloniality and African values — distinctive and intellectually serious.
- NAIRS operational and funding 45 startups before the strategy was even finalised.
- Green AI as a formal objective, ahead of comparable African strategies.
- Honest self-assessment: Nigeria’s own poor global rankings and implementation track record are acknowledged in the document itself.
- Pioneer status for AI infrastructure — a concrete mechanism to attract private capital into the compute gap.
Missing or weak
- No budget. The National AI R&D Fund is proposed with no funding envelope or source attached.
- The governance bodies do not exist. The Ethics Commission and the AI Governance Regulatory Body are both proposed only.
- No hyperscaler region in Nigeria. The entire data-sovereignty vision depends on compute that does not yet exist at scale.
- 34 strategies with no stated sequencing. Without prioritisation, constrained resources scatter across all of them at once.
- Broadband reality gap. National targets exist; Nigeria ranks 140th on broadband readiness.
- The comprehensive AI legislative framework is still not law.
Open questions
What we are watching, and where it stands
Answers re-checked 21 August 2026.
Has the AI Governance Regulatory Body been established and independently staffed?
Not as at August 2026.
Has the AI Ethics Commission been constituted, and who sits on it?
Not as at August 2026.
Has the Digital Economy Bill received assent?
No. Passed the House by May 2026, awaiting the Senate.
When does the National AI R&D Fund open, with committed capital from what source?
No public commitment identified.
Is Pioneer status being granted to data centre and HPC investors?
No published grants identified.
Was the CBN AML roadmap deadline enforced or extended?
Deadline fell around June 2026. No public enforcement action identified either way.
How is NITDA’s dual role managed — implementer of the strategy and promoter of the “independent” regulator?
Unresolved.
What this means for you
If you operate in Nigeria
The strategy is a direction of travel. The regulatory stack is what you will be assessed against, and it is in force today. If you are deploying AI that touches Nigerian personal data, the questions on the table are already concrete: can you evidence human oversight of automated decisions under Section 37, have you run a DPIA for high-risk use, do you know where your Level 3/4 data physically resides, and if you are a bank, did your CBN roadmap go in.
None of those depend on the strategy being implemented. All of them are answerable now, and most organisations we speak to cannot answer them yet.
Primary source: Nigeria National Artificial Intelligence Strategy, September 2025. Regulatory instruments: NDPA 2023 · NDPC GAID 2025 · CBN AML/AI Directive, March 2026 · NITDA National Cloud Policy, October 2025 · NCC Internet Code of Practice, February 2026. Research basis: Board-Level AI Governance Baseline for Sub-Saharan African Enterprises, 2026 edition — Phase 1 regulatory and policy mapping. Penalty figures circulating for the Digital Economy Bill are unverified against primary bill text and are deliberately not reproduced here.
