On Friday 10 April, the Department of Communications and Digital Technologies gazetted the Draft National AI policy for public comment. Please see below commentary by Ahmore Burger-Smidt, Director and Regulatory Head at Werksmans Attorneys for use in any reporting you may be doing on the topic.
South Africa’s draft National Artificial Intelligence Policy is an ambitious and necessary step toward positioning the country within the global AI economy. It shows clear alignment with leading international frameworks, particularly in its adoption of a risk-based approach and its emphasis on ethical AI principles. However, while the vision is compelling, the policy often stops short of providing the level of detail businesses and institutions will need to operate with certainty. This is especially true when it comes to defining risk categories and setting enforceable compliance standards.
Of greater concern is the gap between principle and practice on issues like privacy and data protection. Although the policy signals alignment with POPIA, it does not adequately address how core concepts such as purpose limitation and data minimisation will function in AI systems that rely on large, repurposed datasets. Similarly, the rights of individuals in the context of automated decision-making are underdeveloped, potentially placing South Africa behind more mature jurisdictions like the EU and UK.
The draft also raises questions about implementation. The proposal to introduce multiple new AI-focused institutions including a National AI Commission, an AI Ethics Board, an AI Regulatory Authority, an AI Ombudsperson Office, a National AI Safety Institute, and an AI Insurance Superfund modelled on the Road Accident Fund, designed to compensate people harmed by AI-driven decisions. This could create fragmentation in an already complex regulatory landscape, without clear guidance on roles, coordination, or resourcing.
As the public comment process unfolds, there is a critical opportunity for stakeholders to push for greater clarity, stronger protections, and a more streamlined approach to governance, ensuring that South Africa’s AI ambitions translate into practical, effective regulation.
