It seems hard to recall now, but it was only about four years ago that Artificial Intelligence (AI) was seen as a Silicon Valley curiosity useful to a select few of brilliant data scientists and technologists. But today, thanks to the rapid evolution and widespread adoption of Generative AI (Gen-AI) in particular, AI is no longer a future-facing technology discussion. In South Africa, it is becoming a policy, infrastructure and competitiveness issue.
Much of the public attention over the past few years has focused on generative AI and the speed at which it has entered the mainstream. However, the bigger story is not the technology alone but the way AI is starting to influence how countries regulate innovation, how they invest in infrastructure, and how they think about data as a strategic asset.
In South Africa, at least three shifts are already becoming clear.
1. From fragmented oversight to clearer rules
For some time, AI in South Africa has operated without a dedicated regulatory framework. That does not mean there have been no rules at all. Existing laws such as the Protection of Personal Information Act (POPIA), the Cybercrimes Act and sector-specific regulation have already shaped parts of the environment. But none of these were designed specifically for the realities of AI. So why does progress toward a formal policy framework matter now?
South Africa’s draft National AI Policy Framework, first published for public comment in October 2024, is an important step towards creating clearer direction for business, government and society. If handled well, this can help create the confidence needed for more responsible adoption and investment.
There is often a concern that regulation slows innovation. In practice, the opposite can also be true. Where there is uncertainty, many organisations hesitate. Clearer policy can provide the boundaries within which innovation can happen more confidently. It can help businesses understand what responsible AI looks like, what risks need to be managed, and where accountability sits.
That matters even more in sectors such as financial services, healthcare and the public sector, where trust, compliance and ethics are not optional extras. For South Africa, the real opportunity is to create a practical framework that protects citizens while enabling innovation.
The next phase of AI adoption in South Africa is likely to be less experimental and more operational. As that happens, stronger governance will become essential.
2. AI ambition will be tested by infrastructure reality
The second shift is that AI is no longer just a software story but also an infrastructure story.
In the 2026 Budget Speech, government elevated data infrastructure as a strategic priority and said it would explore options to help data centres and related infrastructure expand investment in South Africa. That is a meaningful signal. It reflects a growing recognition that digital infrastructure is becoming as important to economic competitiveness as more traditional infrastructure.
South Africans know too well that ambition alone will not solve the underlying challenge.
AI and the data centres that support it depend on three things working together: reliable electricity, sustainable access to water for cooling, and high-quality fibre connectivity. South Africa has strengths in digital ambition, but it also has real constraints in each of these areas.
That creates a practical tension. We want to position South Africa as a regional digital hub, but doing so will require infrastructure readiness, not just policy intent.
This is where public-private collaboration becomes important. If South Africa wants to attract and sustain large-scale digital infrastructure investment, it will need to create the conditions for it to happen. That may include faster approvals, more practical partnership models, and greater investment in enabling systems around energy, water and connectivity.
Some of the largest players may be able to solve parts of this challenge through private generation, water efficiency measures and long-term capital investment. But if the market is to broaden, and if more investors are to participate, the ecosystem around them must become easier to navigate.
Whether or not South Africa wants to grow this sector is not in question. But whether we are prepared to build the infrastructure foundation that this growth requires.
3. Data strategy is becoming a national and business issue
The third shift is around data sovereignty and control.
As AI adoption grows, data is becoming more valuable, more contested and more strategic. That means questions around where data is stored, processed, controlled and accessed are becoming more important.
South Africa’s National Data and Cloud Policy already places emphasis on local digital infrastructure and keeping certain categories of government data linked to national security and sovereignty within the country. That signals a broader policy direction towards greater local control over strategically important digital assets.
At the same time, however, these moves sit within a more complex global environment. Governments and large technology markets do not always agree on data localisation, cross-border data flows and digital trade. As these tensions grow, countries such as South Africa will increasingly need to balance openness, competitiveness, security and sovereignty.
For business, this is primarily a strategic issue, not a policy issue.
Organisations will need to become far more deliberate about their data choices. That includes understanding what data they hold, where it sits, which jurisdictions affect it, and what risks come with different operating models. In many cases, a cloud strategy on its own will no longer be enough. Companies will need a clearer data strategy as well.
This will become especially important for organisations operating across borders or in regulated industries, where legal, operational and reputational questions increasingly overlap.
The real opportunity
We know that AI is more than an IT conversation now – it is becoming and has to be a leadership conversation.
For South Africa, the opportunity is significant. But it will not be realised by enthusiasm and intent alone. It will depend on whether we can create practical policy, strengthen enabling infrastructure, and make deliberate decisions about how we govern and use data.
If we get those building blocks right, South Africa can do more than adopt AI. It can shape a more credible, resilient and competitive role for itself in the digital economy.
De Gregorio is a partner in AI Strategy and Enablement at technology and management consultancy iqbusiness.
