The Tortoise and the Algorithm
South Africa's corporate establishment is losing the AI race — and its own behaviour is to blame By Kaveer Beharee, Founder & CEO, Ubiquity AI
There is a game being played across the global economy, and its rules are simple: deploy artificial intelligence faster than your competitors, or watch them pull away. The prize is not incremental productivity. It is structural dominance — the kind that compounds across years and becomes, in time, nearly impossible to reverse.
Singapore understands this. South Africa, by and large, does not.
This is not a polite observation. It is a conclusion drawn from direct experience in both countries, watching how companies respond to the same technological moment from positions of very different resolve. The contrast is not subtle. It is, frankly, disquieting.
The Numbers Tell Part of the Story
The data is unsparing. According to Microsoft’s 2025 AI Diffusion Report, Singapore ranks second globally for AI adoption, with approximately 61% of its working-age population using generative AI tools. The UAE leads at 64%. South Africa, by comparison, ranks fourth among African countries measured — behind Tunisia, Algeria, and Egypt — with a consumer AI adoption rate of just 15%, according to a Cybernews analysis published this month.
These are consumer figures. The corporate picture is, if anything, more revealing. A Morgan Stanley report found that 70% of companies in Singapore have adopted AI. The Oxford Insights Government AI Readiness Index ranks Singapore third in the world, with a score of 80.79 out of 100. South Africa does not appear in the top tier. Across the African continent, 11 of the 12 least AI-prepared countries in the world are located there, a fact that ought to concentrate minds considerably.
Numbers, of course, obscure as much as they illuminate. The more instructive question is not what the figures show, but why the gap exists — and whether it is narrowing or widening. The evidence suggests the latter.
A Tale of Two Philosophies
The difference between Singapore and South Africa is not primarily one of resources, talent, or technological access. It is one of corporate philosophy — specifically, of how each country’s business establishment has answered a single question: should we build AI ourselves, or partner with those who already have?
Singapore answered that question decisively, and early. The city-state launched its first National AI Strategy in 2019 — before ChatGPT existed, before generative AI became a dinner-table conversation. By the time the boom arrived, Singapore’s institutions were already primed to absorb it. The government committed more than S$1 billion over five years to AI research and adoption. It deployed AI in schools, hospitals, immigration, and public services. It built structured, funded pipelines connecting enterprises directly to startups at the cutting edge. Enterprise Singapore, the government’s business development agency, now co-funds AI Centres of Excellence across industries and has partnered with Google Cloud, Microsoft, and Amazon Web Services to accelerate deployment at scale.
The result is a corporate culture that treats AI as a utility rather than a trophy. Singaporean companies, having long understood their domain expertise, are secure enough to acknowledge what they do not know. They do not insist on building the tools themselves. They partner — deliberately, quickly, and at pace. Tech partners are appointed after one or two presentations. Pilots begin within weeks. The distance from a first meeting to a live deployment is measured in months, not years.
South Africa’s experience, as seen from the inside, is rather different.
Innovation Theatre
One of South Africa’s four largest banks — an institution managing hundreds of billions in assets, serving millions of customers, employing thousands of people — has recently all but disbanded the innovation team that operated at group level. The team has been quietly reintegrated across the business. The innovation centre: closed. The experiment: over.
This is not an isolated case. It is a symptom.
Visit enough corporate innovation centres in South Africa, attend enough industry conferences — particularly in financial services — and a pattern emerges with uncomfortable clarity. The language is fluent. The buzzwords are current. The PowerPoints are polished. But beneath the vocabulary of transformation, there is, in too many cases, very little substance — and almost no deployment.
The AI initiatives one encounters locally tend to fall into one of two categories. The first is driven by fear of missing out: a competitor announced something, the board asked a question, and now the organisation must be seen to respond. The second is designed to manufacture a reassuring narrative for shareholders — evidence, however thin, that the company is modern, forward-looking, and engaged with the future. Neither category produces competitive advantage. Both consume resources that could be deployed more productively elsewhere.
The structural cause is not difficult to identify. South African corporates are deeply resistant to the idea of sharing intellectual property with the technology partners — and especially the young startups — who are building the most powerful tools. The instinct is understandable. The consequences are damaging. Rather than partnering with the firms at the frontier, large organisations default to building in-house capabilities using teams that lack the specialisation the moment demands, or to procuring from legacy technology vendors whose innovation credentials are barely stronger than their own.
Death by Governance
There is another force at work, and it is perhaps the most corrosive of all: process.
In the course of building Ubiquity AI, this writer has been involved in engagements where a business owner — motivated, capable, genuinely convinced of the value of the technology — spent months attempting to secure a meeting with their own organisation’s technology team. Several more months passed before corporate governance could be brought into the discussion. Then came the legal review. Then the NDA. By the time the paperwork was finalised, the competitive window had moved.
This is not an anecdote about bureaucracy. It is a description of a culture that has structurally misaligned its decision-making with the speed at which the technology is moving. Governance, in this formulation, has become an instrument of delay rather than a mechanism for responsible deployment. The outcome is that the organisations most in need of transformation are the ones least capable of achieving it quickly.
Singapore offers an instructive counterpoint. Under its Enterprise Compute Initiative, launched in 2024, the government co-funds AI Centres of Excellence, provides up to S$250,000 in cloud credits per business, and offers co-investment in consulting costs for companies developing AI roadmaps. In August 2025, Microsoft and Digital Industry Singapore launched an Agentic AI Accelerator offering 300 businesses structured support to move from strategy to deployment within twelve months. The program does not ask companies to build the technology themselves. It asks them to commit to using it.
The contrast in velocity is stark. In Singapore, companies are known to appoint technology partners within a week of a credible presentation and be in live deployment within weeks thereafter. The city-state’s AI startup ecosystem has collectively raised $8.4 billion in venture capital — a figure that dwarfs its regional peers and signals the seriousness with which commercial AI is taken as an economic priority.
The Productivity Gap Looms
Zooming out, the stakes are higher than any individual sector or firm. McKinsey’s 2025 Global Survey found that 78% of companies globally now use AI in at least one business function — nearly four times the figure from 2017. Generative AI adoption more than doubled between 2023 and 2024 alone. The global AI market is projected to reach $1.81 trillion by 2030. PwC estimates AI could add up to $15.7 trillion to world GDP over the same period.
Countries and companies that are compounding their AI capabilities now will command structural advantages that are difficult to reverse. Those that are still debating governance frameworks while their counterparts ship products are not simply falling behind — they are making their eventual catch-up exponentially more expensive.
South Africa’s position is particularly precarious. It is an economy already contending with constrained growth, elevated unemployment, and a financial services sector facing significant IFRS-9 impairment pressure. AI represents one of the most credible available mechanisms for improving credit risk management, reducing operating costs, expanding financial inclusion, and rebuilding institutional resilience. The tools exist. The talent exists. What appears to be missing is the organisational will to move.
What Would Change This
There is a version of this story that ends differently, and it is not fantastical.
South Africa has genuine strengths: deep financial services expertise, a regulatory environment that, for all its complexity, is more sophisticated than most of the continent, and a generation of technology entrepreneurs who are building world-class products despite — not because of — the corporate establishment’s reluctance to engage them. The irony is considerable. Some of the most innovative AI infrastructure being built for African financial services is being developed by South African startups, several of which are finding their earliest serious commercial traction not in Johannesburg or Cape Town, but in London, Paris, and Singapore.
The prescription is neither complicated nor particularly expensive. It requires, first, that large organisations accept the fundamental insight that Singapore’s corporate culture has already internalised: domain expertise is the asset; AI is the tool; and the fastest path to competitive advantage runs through partnership, not in-house construction. It requires, second, that governance be redesigned not as a barrier to deployment but as a framework for responsible acceleration. And it requires, third, that corporate South Africa begin to treat its domestic technology startups as strategic assets rather than vendors to be managed at arm’s length.
The window is not closed. But it is not open indefinitely.
By the time South Africa finishes the paperwork, Singapore has shipped the product, onboarded the customers, and moved on to version two. That sentence, uncomfortable as it is to write, is not hyperbole. It is a description of the current state of play — and a warning about what becomes permanent if the pace does not change.
Kaveer Beharee is the Founder and CEO of Ubiquity AI, a predictive AI company solving the $7.4B global collections inefficiency. He was recognised as a Young Enterprise Initiative Laureate in France (2022) within the VivaTechnology ecosystem, and previously advised Parliament of South Africa on the National Credit Amendment Act. He writes on AI, financial services, social justice, financial inclusion and emerging market competitiveness.
Follow on LinkedIn: linkedin.com/in/kaveer | Twitter/X: @UbiquityAI