South Africa Chooses AI Coordination Over Central Control

South Africa published its draft AI policy April 2, 2026, taking a unique approach by spreading responsibility across existing agencies instead of creating a super-regulator. Unlike Nigeria and Kenya which are moving toward centralized AI oversight, South Africa favors coordination through existing regulators like the Financial Sector Conduct Authority for financial AI and SAHPRA for medical AI systems. The policy, fully expected by 2027/2028, uses risk-tiered regulation with four categories: unacceptable (banned), high (strict scrutiny), limited (lighter rules), and minimal oversight. This approach argues that sector-specific regulators already have the relevant expertise. However, the framework creates an enforcement gap - coordinating bodies can guide but cannot compel action, potentially leading to uneven oversight across different sectors. For businesses, this means uncertainty as AI systems evolve and may face changing compliance requirements. South Africa also emphasizes building local datasets and African language processing to create more inclusive AI systems, though this adds complexity to data governance. For Nigeria, considering its own approach to AI regulation, this model offers an alternative to centralized control but highlights the challenge of ensuring consistent enforcement across different regulatory bodies with varying capacities.

SOURCE: https://techcabal.com/2026/04/09/south-africas-new-policy-is-betting-on-decentralised-ai-control/