As South Africa prepares to host the G20 Summit in 2025, experts say the moment is ripe for the country to take a leading role in shaping global artificial intelligence (AI) governance – particularly around transparency, accountability, and human oversight.
The call comes just months after the European Union’s landmark Artificial Intelligence Act came into force in July 2024 and shortly after South Africa released its own National AI Policy Framework in October. The framework outlines how policymakers aim to use AI to promote social and economic prosperity, and public consultations are now underway.
However, experts say South Africa urgently needs local regulations that address the risks of AI in a realistic, context-specific, and enforceable manner. Without clear legal guardrails, the country risks falling behind regional peers like Nigeria, Rwanda, and Mauritius, which are actively developing their own AI strategies.
President Cyril Ramaphosa established a Presidential Commission on the Fourth Industrial Revolution in 2020, predicting that by 2030, the country would have “fully harnessed the potential of technological innovation to grow our economy and uplift our people.” But meaningful governance is now key to delivering on that vision.
Africa’s informal data landscape presents unique challenges. Much of the continent’s data is stored in servers overseas, raising concerns over data ownership, sovereignty, and digital colonialism. South Africa’s Protection of Personal Information Act (POPIA) offers some protection, but broader legal infrastructure is needed—especially as global powers compete for access to Africa’s data as a resource.
Rwanda has responded by declaring open data – including public records and financial information—a national asset under its Vision 2025 strategy. The aim is to localize data storage and ensure national control over strategic information.
Continental projections are optimistic: Africa’s AI-driven innovation is expected to generate $1.2 trillion in economic value by 2030, contributing 5.6% to GDP. But much will depend on how governments manage risks tied to bias, surveillance, misinformation, and the weaponization of AI against vulnerable groups.
Human rights advocates warn that AI tools – often trained on data and algorithms developed outside of Africa – may not reflect local realities and could reinforce racial, class, and cultural biases. There are also growing fears that AI systems could be used to distort political narratives, suppress dissent, or enable disinformation campaigns.
Examples abound: In Sudan, AI-generated audio mimicking military officials led to confusion and fear online. During South Africa’s 2024 election, a deepfake video featuring Donald Trump appeared to endorse the newly formed MK Party. Though easily debunked, experts say future disinformation could be harder to detect and more damaging.
In Rwanda, critics allege that AI-powered tools are being used to generate vast quantities of state-sponsored content aimed at drowning out dissenting voices and manipulating social media platforms’ engagement metrics.
“AI tools are not neutral,” says Guy Berger, a South African-based expert on internet and media policy. “They carry cultural, algorithmic, and linguistic biases. We can’t ignore the origins of these technologies, especially with most coming from the U.S.”
To ensure AI benefits rather than harms African democracies, Berger and others argue that countries need access to training data relevant to their own social, linguistic, and political contexts. Universities, governments, and civil society must collaborate to support homegrown innovation and research.
In February 2025, the EU will enact new restrictions, banning certain AI practices like predictive policing and biometric identification in public spaces. These regulations could serve as a model for African states looking to assert digital sovereignty and protect citizens.
The upcoming G20 Summit offers South Africa a global platform to drive these conversations. Brazil, host of the 2024 summit, made AI governance a central theme. Now, it’s South Africa’s turn to push the dialogue forward.
“This is a critical moment for Africa to assert its voice in global tech policy,” Berger said. “We need to make sure AI works for our people – not the other way around.”
