When AI governance is still being defined at a national level, having control at an organisational level separates businesses from the competition.
When South Africa’s government suddenly withdrew its newly gazetted national AI policy framework, the document itself was flagged with fictitious sources. Ironically, these sources were likely hallucinated by the very technology the policy sought to govern. This was a very uncomfortable AI moment for us, but also a teaching one.
This isn’t just a case of AI still being in its infancy. It proves that human judgement is what will drive success in this AI-centric world.
South Africa finds itself at a defining inflection point. According to Microsoft’s Global AI Diffusion report, AI adoption is skyrocketing, with local usage reaching 23.1% in the first quarter of 2026, placing the country in 46th place out of 147 economies.
While that progress is encouraging, the same report shows that the gap between the Global North and the Global South is widening. Adoption is accelerating more than twice as fast in higher-income regions. The challenge is not a lack of enthusiasm. It comes down to infrastructure, skills and the kind of deliberate, leadership-driven commitment from local businesses that turns experimentation into impact.
Stuck in the pilot trap
Across Africa, the AI conversation has evolved. The question is no longer whether to adopt the technology, but rather how businesses can move beyond the pilot phase and deploy AI at scale.
PwC Africa’s latest AI research found that just 20% of companies are capturing 74% of AI-driven value. The finding points to a widening gap between businesses experimenting with AI and those successfully embedding it into their operations at scale. Most businesses are testing AI at the edges rather than embedding it into their operations.
The cost of hesitation is not hypothetical. Every major technological advance in the last century rewarded businesses willing to adapt early while leaving hesitant organisations behind. Organisations that fail to fundamentally reinvent their business models face strategic irrelevance within a decade. Even if the strategy is simply to integrate AI-enabled applications into existing workflows, that strategy still demands a total rethink.
The regulation gap and the governance imperative
South Africa’s national AI policy won’t be finalised until the 2026-2027 financial year. When it arrives, it will most likely take a sector-specific, risk-based approach before being layered onto existing legislation.
Currently, the Protection of Personal Information Act (POPIA) is one example of existing South African legal frameworks that may apply to AI. This is mainly because of one specific section covering automated decision-making. Still, waiting shouldn’t be an option.
Putting your business’s AI adoption plans on hold due to incoming regulations is a quick road to competitive disadvantage. Smarter businesses are already setting up internal governance structures as a matter of practice. No legal obligation is forcing businesses to do this. The time leading up to new policies being enacted is an opportunity to define standards rather than seek compliance.
The human advantage as competitive advantage
Microsoft’s Global AI Diffusion report reveals Africa’s developer community is growing at 21% annually, the fastest regional rate in the world. South Africa’s more than 500,000 developers represent one of the continent’s most mature technology ecosystems.
Human talent is a legitimate competitive asset. It should be amplified by AI, not replaced by it. Enterprises that get the most from their AI investments are the ones who refocus their people on higher-value work where interpretation, contextualisation, creative problem-solving and nuanced judgement are more often needed, and cannot easily be replicated by AI models.
This is what ‘human advantage’ means in practice. First, you need to recognise AI as a force multiplier for human capability. Then you need to invest in the skills, culture and leadership that harmonises people and technology.
Smaller and mid-sized businesses may have a particular edge here. Without the legacy systems and approval structures that slow larger organisations, agile enterprises can move faster, learn quicker and keep pace with this rapid evolution of tech.
Building on the right foundation
For South African enterprises, the hardware and software underpinning AI deployment matters more than it might appear. Latency, security, processing capability and device-level management all affect the performance of AI tools.
At ASUS, we address this through a portfolio of commercial-grade solutions built for enterprise AI workloads. We combine the processing capability required for AI-assisted workflows with layered security through platforms like ASUS ExpertGuardian. This kind of solution enables IT teams to monitor, manage and protect devices across hardware, firmware and application levels.
South Africa’s AI story is still being written. The regulatory chapter is incomplete; the infrastructure gaps are real and the divide between ambition and execution remains a gap many organisations are still hesitant to cross.
What is clear is that the businesses who lead are those that treat AI not as a technology decision but as a strategic one anchored in human capability. That’s how they take AI from experiment to production, and how they will start to generate real value from their investments.
