Webinar | Session 2: Data Governance, Privacy and Cross-Border Flows

Webinar | Session 2: Data Governance, Privacy and Cross-Border Flows

🎙 Mandela Institute 👥 102 📅 May 25, 2026 ⏱ 108 min 👁 10 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

data sovereigntyPOPIAAI regulationinformation regulatorcross-border data flows

Summary

This webinar, part of a series on South Africa’s AI policy, focuses on data governance, privacy, and cross-border data flows. Professor Sylvia Papadopoulos argues that AI governance requires a coherent data governance architecture, emphasizing the need to translate POPIA principles into the AI lifecycle. She critiques the withdrawn draft AI policy for lacking institutional clarity and proposes a single AI-specific statutory authority to coordinate across regulators, rather than creating multiple overlapping bodies. Advocate Lebogang Stroom, from the Information Regulator, outlines the regulator’s dual mandate under POPIA and PAIA, stressing the importance of balancing privacy with information access. She discusses the regulator’s role in AI-related personal information processing and the need for a risk-based approach. Siyabonga Skosana from Samsung provides an industry perspective, highlighting practical challenges and the need for clear regulatory frameworks to enable innovation while ensuring compliance. The discussion covers data sovereignty, cross-border transfers, open data, and the role of procurement in enforcement. The speakers agree on the need for a coordinated, risk-based approach that leverages existing laws while addressing gaps. The Q&A session addresses audience questions on enforcement, international cooperation, and the practical implications for businesses.

188 words

Critical Evaluation

Value of the Information & Strength of the Argument

The webinar provides substantial value by offering a multi-stakeholder perspective on AI data governance in South Africa. Professor Papadopoulos delivers a well-structured argument for institutional coherence, using the withdrawn draft policy as a case study. She effectively distinguishes between horizontal and vertical regulation and proposes a pragmatic model of a single coordinating authority. Advocate Stroom provides authoritative insights into the Information Regulator’s mandate and enforcement approach, grounding the discussion in existing legal frameworks. Mr. Skosana offers a practical industry viewpoint, highlighting the need for regulatory clarity to foster innovation. The argumentation is generally solid, with speakers building on each other’s points and addressing counterarguments. However, the discussion is largely policy-oriented and lacks empirical evidence or case studies to support some claims. The value lies in its expert analysis and actionable recommendations for policymakers.

Scientific Rigor, Source Quality, Title Accuracy

The webinar demonstrates a high level of scientific rigor in its reliance on existing legal frameworks, particularly POPIA and PAIA. The speakers are recognized experts in their fields, and their arguments are grounded in legal analysis and practical experience. However, the webinar does not cite specific academic sources or empirical studies, relying instead on expert opinion and legal interpretation. The title accurately reflects the content, focusing on data governance, privacy, and cross-border flows. The discussion is well-structured and stays on topic. The absence of citations is a minor weakness, but the overall rigor is high given the nature of the webinar as a policy dialogue.

253 words

Title / Content Match

The title accurately reflects the session's focus on data governance, privacy, and cross-border flows in the context of AI policy.

Quality & Reliability

8/10

The webinar features three expert speakers with relevant academic, regulatory, and industry backgrounds. The discussion is structured and grounded in existing legal frameworks (e.g., POPIA), with clear arguments and practical recommendations. However, it is a policy dialogue rather than a peer-reviewed study, and some claims are not backed by specific citations.

Key Moments

Cited Sources

Concurring Sources

  • OECD AI Principles — Aligns with the webinar's emphasis on risk-based approaches and human oversight.
  • Information Regulator South Africa — Official website of the regulator, providing context on its mandate and activities.

Dissenting Sources

  • None — No discordant sources were identified in the webinar.

Contribution & Novelties

The webinar contributes to the ongoing policy debate on AI governance in South Africa by offering a nuanced analysis of the institutional and legal challenges. It proposes a concrete model for a single AI-specific statutory authority to coordinate across existing regulators, which is a novel approach in the South African context. The discussion also clarifies the concept of data sovereignty, advocating for enforceable control rather than crude localization, and emphasizes the need to operationalize POPIA principles across the AI lifecycle. The inclusion of perspectives from academia, the regulator, and industry provides a comprehensive view of the issues.

Pour aller plus loin :

149 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, indicating a content-rich and well-structured discussion. The technical level is moderately high, reflecting the legal and policy focus. The overall reliability is strong, given the expertise of the speakers and the grounding in existing legal frameworks.

Reliability 8/10

💬 No comments were provided for analysis.