BRICS Summit on AI Governance & Future AI Policy Development: My Experience as South Africa’s Representative
- Donnavan Linley
- 4 days ago
- 3 min read
In July 2025, I attended the pre-BRICS summit hosted by the Brazilian Development Banks (BNDES) in the beautiful city of Rio de Janeiro focusing on ‘Governance and Public Strategies in AI’. Attending the summit was an invaluable opportunity to network and understand how other nations are approaching AI policy and developing AI capabilities. It's clear that countries are prioritizing the development of their own national AI sovereignty. Thus, highlighting the need for African nations to establish their own policies and strategies in building AI capabilities.
The summit was opened by high-ranking Brazilian officials including several Brazilian ministers, the Mayor of Rio de Janeiro and the President of BNDES.
I participated in a discussion on the ‘Data Economy and AI governance’ where I presented the South African perspective. My presentation in the panel discussion highlighted five major considerations central to effective regulation, AI-related issues and opportunity creation within the fast-evolving landscape of artificial intelligence.
Competition Policy and Market Inquiries
A primary focus was on competition policy, specifically the use of market inquiries as a regulatory tool in addressing AI's impact on markets and consumers. With AI capabilities and AI-embedded digital platforms in constant evolution, market inquiries provide a flexible, forward-looking, evidence-led mechanism that involves stakeholders in seeking practical solutions. This approach supports the design of remedies for adverse effects AI might introduce, without constraining the innovation and benefits that AI products and services may offer.
Access to Data
Issues around access to data were traversed, especially in training AI models such as large language models (“LLMs”) which embody economies of scope and scale for data usage. This creates high barriers to expansion for SMEs and startups facing significant challenges accessing essential data to develop their AI models.
A "data rich, data poor" divide is widening globally. While countries like the USA are rapidly accelerating their data collection capabilities, many emerging nations are lagging, expanding theirs at a much slower rate. This asymmetry is also evident between big tech platforms and local South African companies. For AI chatbots to be relevant and responsive in a local context, it's crucial to include and share South African data in the development of LLMs.
Access to AI Infrastructure
Building the necessary digital infrastructure for AI, such as data centres, supercomputing, and high-speed networks, is a capital-intensive endeavour. In South Africa, while we have our own data centres like Teraco, we also rely on foreign-owned ones from giants like AWS and Microsoft. This often leaves local small and medium-sized enterprises (SMEs) and startups facing potentially higher costs and less localized support when they use cloud services.
A significant investment gap is growing between leading AI players like the US, China, and big tech, and many emerging countries. This is where competition policy becomes crucial. It can ensure that access to essential AI infrastructure, such as cloud services, is fair, non-discriminatory, and non-exploitative. Furthermore, expanding digital services and infrastructure into underserved rural and poorer areas is essential for achieving digital inclusion.
Legal Jurisdiction
The importance of nation-states in establishing legal jurisdiction to assess and regulate global technology platforms was a key point of discussion.
Opportunities for Domestic AI Development
While developing and securing AI sovereignty is important, developing local AI capabilities reveals both niche and industry-specific opportunities, for example, in sectors such as agriculture, education, and healthcare. Opportunities also exist in creating AI applications and developing large language models (LLMs) in local South African languages.
Brazil's commitment to AI opportunity creation is evident in its plan to invest $4 billion in developing its own AI capabilities. The summit's structure offered a clear roadmap for this kind of development and African countries can benefit from the summit’s key focus areas:
Building a robust data economy and effective AI governance.
Integrating AI into government services.
Developing sustainable infrastructure, such as 'green' data centres, to address AI's high energy consumption.
Creating AI hubs (like Rio de Janeiro) to attract talent, establish data centres, and foster collaboration.
Exploring private sector initiatives and potential innovations
Author's Bio
Donnavan Linley is currently Senior Analyst and the Technical Lead for the Media and Digital Platforms Market Inquiry (“MDPMI”) at the Competition Commission of South Africa. With over a decade of experience in competition economics, he currently specializes in digital markets and AI. In his role, he assesses features of industries such as generative AI, search, and AdTech to identify potential competition issues and the impact of platform behaviour on the news media industry. Previously, he was the team lead for the travel accommodation and search streams of the Online Intermediation Platform Market Inquiry (OIPMI). His work comprises of rigorous investigations, research and analysis to solve complex problems and innovate solutions to resolve competition issues and market failures.