The missing map: Africa, the Gulf, and Latin America beyond Brazil
Lesson 4 of 5 in The UK, Canada, Brazil — and the Missing Map: Africa, the Middle East, Latin America.
Most AI-governance curricula draw a map with perhaps ten countries on it and call it ‘global’. That map is wrong twice over. It misses jurisdictions making real governance moves — and it misses that most of humanity lives in places whose AI future is being decided largely by rules written elsewhere. For a working analyst, the missing map matters for a hard commercial reason too: your clients deploy there.
Start with the continent most often skipped entirely. In July 2024 the African Union Executive Council endorsed the AU Continental AI Strategy — a five-year framework for all 55 member states built around harnessing AI for development (Agenda 2063), building African compute, data and skills capacity, and asserting data sovereignty: the principle that African data should generate African value rather than being exported raw and re-imported as someone else’s product. It is soft law — the AU cannot bind member states the way Brussels binds EU members — but it does what continental soft law does: sets shared vocabulary, legitimises national action, and coordinates negotiating positions in UN fora, where African states pushed the capacity-building agenda hard.
Beneath the AU sits Smart Africa, a Kigali-headquartered alliance of 40 or so states that produced the AI for Africa Blueprint (2021) under Rwanda’s leadership — the practical playbook many national strategies drew on.
African pioneers
Mauritius published Africa’s first national AI strategy back in 2018. Egypt followed in 2021 (refreshed mid-decade), and South Africa issued a National AI Policy Framework in 2024.
Nigeria released its National AI Strategy in August 2024 — drafted through an unusually open process co-led by the digital economy ministry and researchers, anchored by NITDA and its AI and robotics centre, and sitting on top of the Nigeria Data Protection Act 2023.
Kenya launched its National AI Strategy 2025–2030 in March 2025, building on its 2019 Data Protection Act — and became the first African state to join the International Network of AI Safety Institutes. A cautionary tale sits alongside: a 2023 private-member’s bill to license AI practitioners through a statutory ‘robotics society’ was so unworkable it united the entire tech sector against it — a reminder that bad first-mover statutes are worse than strategies.
Rwanda adopted its National AI Policy in 2023, co-developed with Smart Africa, and in April 2025 hosted the Global AI Summit on Africa in Kigali, where dozens of governments backed an Africa Declaration on AI calling for African compute, African data infrastructure, and an African voice in global rule-making.
The Gulf
The Gulf states run a third model: neither the EU’s rulebook nor the UK’s regulator network, but state-led AI industrial strategy with governance attached.
The UAE appointed the world’s first Minister of State for AI (Omar Al Olama) in October 2017, set a National AI Strategy 2031, released open-weight Falcon models through Abu Dhabi’s Technology Innovation Institute, and in 2024 issued a UAE Charter for the Development and Use of AI — principles on safety, privacy, fairness and human oversight. It will host the 2028 global AI summit. Binding horizontal AI law: none — the bet is that being the friendliest serious jurisdiction wins the industry.
Saudi Arabia concentrated everything in SDAIA — the Saudi Data and AI Authority (2019) — which is simultaneously data-protection regulator, national AI strategist and flagship-project builder. SDAIA issued AI Ethics Principles (adoption expected for government entities, encouraged for the private sector) and generative-AI guidelines for officials and the public. Watch the structural feature, not the documents: promoter and regulator are the same body — the exact conflict Canada’s critics attacked in ISED, here by design.
Latin America beyond Brazil
Chile is the region’s quiet sophisticate: a national AI policy since 2021 (updated 2024), UNESCO’s readiness methodology piloted early, and since May 2024 a draft AI bill consciously modelled on the EU’s risk-based architecture — unacceptable-risk bans, high-risk obligations, an enforcement agency. Chile also convened the Santiago Declaration process (October 2023), where twenty-plus Latin American and Caribbean states proposed a regional intergovernmental AI council.
Peru did something nobody else in the region had done: it passed an AI statute — Law 31814 (July 2023), a short promotion law declaring AI development of national interest with human-rights guardrails, closer in spirit to Japan’s promotion act than to the EU Act. Proof that ‘first law’ and ‘strongest law’ are different races.
Mexico is the region’s cautionary counterpoint: dozens of AI bills introduced in Congress since 2020, an early national strategy (2018) that lapsed with a change of government, and — as of September 2026 — no adopted national AI framework. Bill introduction is cheap; Mexico teaches you to distinguish legislative noise from legislative progress.
Interactive sorting exercise: Status check: sort each jurisdiction by where its AI governance actually stands (as of September 2026).
Tool: Global Governance Atlas — Open the Governance Atlas and walk the whole map — every jurisdiction in this lesson, colour-coded by instrument type and status.
Key terms: au continental ai strategy, smart africa, data sovereignty, sdaia, soft law
Interactive checkpoint quiz (2 questions) — open this page in a browser to take it.