AU Continental AI Strategy Phase 1: How African Member States Are Building National AI Strategies in 2026

What the Continental AI Strategy Actually Is

The African Union Continental AI Strategy, endorsed in July 2024, is the AU’s first dedicated framework for AI governance and adoption across all 55 member states. It is structured as a 2025-2030 implementation plan with two phases: Phase 1 (2025-2026) focused on foundations — national strategies, governance structures, resource mobilization — and Phase 2 (2028 onwards) focused on executing core projects after a 2027 review.

According to analysis by the Future of Privacy Forum, the Strategy proposes five core governance activities: amending existing laws on data protection, cybersecurity and consumer protection; identifying regulatory gaps in labor protections and algorithmic bias; establishing enabling national AI strategy frameworks; developing assessment tools including impact assessments and independent review mechanisms; and continuous African-led research to evaluate governance effectiveness.

The Strategy is not a binding regulation. It is a coordination instrument — closer in spirit to the EU’s AI Act guidance for member states than to a directly applicable rulebook. Its leverage comes from peer pressure, capacity-building support, and the AU Data Policy Framework with which it explicitly integrates.

What Phase 1 Actually Demands of Member States

The 2025-2026 deliverables for member states fall into four categories.

National AI strategy. Every member state is expected to publish a national AI strategy that “domesticates” the continental framework — adapting it to the country’s economic structure, sectoral priorities, and existing legal stack. The strategies are expected to cover priority sectors (health, agriculture, education, public administration), data governance, talent development, and infrastructure (compute, connectivity, sovereign cloud).

Governance structures. Each member state is expected to create or designate the institutional body responsible for AI policy coordination — an AI advisory board, an AI office within an existing ministry, or a national AI authority. The Strategy also calls for a continental-level Advisory Board on AI and a regional AI Ethics Board.

Resource mobilization. Phase 1 explicitly includes mobilizing financial and technical resources, including from international partners (UNESCO, the World Bank, the African Development Bank) and private partners. The Strategy does not set a continental budget figure; resource mobilization is country-by-country.

Centers of excellence and capacity building. Phase 1 calls for the establishment of national or regional centers of excellence to anchor research, training, and applied AI work.

Country-by-Country: Where the Domestication Stands

The pace of national-level adoption has been uneven, but the trajectory in 2026 is clear.

Adopted strategies. According to the Future of Privacy Forum analysis and tracking by White & Case, six AU member states have published national AI strategies: Algeria, Benin, Egypt, Mauritius, Nigeria, and Senegal. Rwanda also has a national AI policy, published in 2023, that pre-dates the AU Strategy and has served as one of the de facto reference templates.

In development. A second wave of countries is drafting national strategies: Ethiopia, Ghana, Kenya, Morocco, South Africa, Mauritania, Tanzania, and Tunisia. Kenya formally introduced its National AI Strategy (2025-2030) in March 2025 and committed approximately KES 152 billion (around $1.14 billion) over five years — the largest public commitment to AI on the continent so far.

Country anchors. Morocco’s Mohammed VI Polytechnic University hosts an International Centre on Artificial Intelligence designated as a UNESCO Category II center in November 2023 — one of the institutional anchors for AU regional capacity building. Egypt’s strategy emphasizes Arabic-language AI and public-sector applications. Senegal’s strategy is the most explicit on the data-sovereignty argument.

The pace gap. The half of AU member states that have neither adopted nor publicly drafted a national strategy is the most strategically important question for Phase 1. The AU’s continental ambition rests on whether countries with limited public-sector capacity — small island states, post-conflict states, low-income economies — can produce credible national strategies by the 2027 review or whether the continental framework effectively becomes a two-speed Africa.

What Phase 1 Implementation Looks Like in Practice

For African policymakers, multinationals operating across the continent, and international partners, the practical question is how to engage with Phase 1 in 2026 rather than wait for Phase 2.

1. Treat the AU Data Policy Framework as the joint operating system

The Continental AI Strategy is explicitly designed to sit on top of the AU Data Policy Framework. Member states that already have credible data protection laws (Algeria’s Law 18-07 amended by Law 25-11, Kenya’s Data Protection Act 2019, Nigeria’s Data Protection Act 2023, the Mauritius Data Protection Act 2017) have a working substrate. Countries without one need to ratify the Malabo Convention or pass equivalent national legislation before AI governance can be enforced. International partners advising member states should treat the data law gap as the binding constraint — without it, the AI strategy is aspirational.

2. Anchor the national strategy in three named sectors, not all of them

The countries with workable strategies (Rwanda, Kenya, Egypt) have been disciplined about sectoral focus. Trying to cover health, agriculture, education, finance, manufacturing, transport, justice, and public administration in a single national strategy produces a document with everything and a budget for nothing. Strategies that name three priority sectors with explicit resource lines outperform comprehensive but unfunded ones. The 2027 review will reward delivery, not breadth.

3. Build the talent pipeline before the compute infrastructure

The instinct in many member-state strategies is to lead with a sovereign AI compute commitment — a national GPU cluster, a sovereign cloud, a foundation model. The talent gap is the harder constraint. Africa graduates roughly 700,000 STEM graduates per year against demand that is growing faster, and AI-specialized talent is concentrated in 5-6 countries. Phase 1 success looks like a coordinated talent program (university curriculum reform, scholarship pipelines, returning-diaspora packages) anchored to the centers of excellence the Strategy calls for. Compute without talent is idle racks.

4. Engage continental and global partners on a portfolio basis

UNESCO, the African Development Bank, the World Bank, the Smart Africa Alliance, and several bilateral donors are all active on AI in Africa. Member states that engage them on a portfolio basis — with a clear national strategy as the coordination document — secure better-aligned support than countries that take ad-hoc grants. The AU’s own communications describe AI as a strategic priority for socioeconomic transformation; the financing follows the strategies that read as serious operational documents, not policy statements.

What Comes Next in 2027

Phase 1 ends with a review in 2027. That review will be the first formal scorecard of how seriously member states have taken the continental framework — how many national strategies are adopted, how many governance bodies are operational, how much financing has been mobilized, and what early use cases have moved from pilot to production.

The realistic outcomes are three. The strong case is that 25-30 member states have adopted national strategies and the AU has a credible continental advisory board, regional ethics board, and pipeline of cross-border use cases (cross-border health AI, AfCFTA-related logistics AI, regional disaster response AI). The middle case is that 15-20 strategies exist on paper but most lack budgets and operational structures, and the 2028 Phase 2 launch is delayed. The weak case is a two-speed Africa: a handful of leading countries (Egypt, Kenya, Morocco, Nigeria, Rwanda, South Africa) execute meaningfully while the majority remain in scoping. Each scenario has different implications for international partners, multinational technology firms, and African startups planning continental expansion.

The Continental AI Strategy is the most ambitious AU policy instrument since the AfCFTA. Its success or failure in Phase 1 determines whether African AI governance becomes a coordinated project or a collection of national experiments — and whether the continent enters Phase 2 as a single market for AI policy or as fragmented national markets that companies must navigate one by one.

Frequently Asked Questions

Sources & Further Reading

• Continental Artificial Intelligence Strategy — African Union
• The African Union’s Continental AI Strategy — Future of Privacy Forum
• AI Watch: Global regulatory tracker — African Union — White & Case
• African Union Champions AI for Africa’s Socioeconomic Transformation — AU Press Release
• Communique: High-Level Policy Dialogue on AI in Africa — African Union
• UNESCO’s support to AU’s Continental AI Framework and Strategy — UNESCO