The following is a summary of my keynote given on 28th November 2025 at the Platinum Zambia 1000 Summit.
My theme today is simple but uncomfortable: are we actually educated for the next frontier?
To answer that honestly, we first have to ask two questions we almost never ask. What is education for? And what exactly is the frontier?
Historically, in Africa, we have not asked those questions seriously. We have mostly adapted to someone else’s answers.
Our ancestors didn’t ask, “What is the global frontier?” They were trying to survive and govern in their own context. The founders of our nations also didn’t really ask, “What is the frontier of the twenty-first century?” They were trying to catch up with the twentieth.
In each era, our education systems have been built to adapt, not to define the frontier.
If we go back: before colonialism, education in our societies existed to govern, to produce, to defend, to conquer. You learned how to blacksmith, how to brew, how to make clothes, how to make medicine, how to be a warrior, how to farm. These skills weren’t random. They emerged from an Iron Age context. And because Bantu societies had those skills, they displaced earlier hunter–gatherer groups like the San. This pattern was global: those who mastered the skills that defined an age tended to dominate those who did not.
Then Europeans arrived on our coasts.....
For about three hundred years, they largely traded with African polities before pushing aggressively into the interior. Those three centuries were not idle. They spent that time exploring the coastline, mapping rivers, cataloguing resources, building blue-water fleets, inventing instruments, and developing the financial and legal tools for oceanic empire.
We, on the other hand, assumed the world shared our values about reciprocity and trade. We did not send our own expeditions to understand Europe. We did not build our own blue-water shipping. We did not build our own science institutions at scale. We adapted to what we thought the present was, while they were quietly building the infrastructure of a future in which we would be the raw material.
So what we Bantu once did to the San, Europeans then did to us. They arrived with the skills and institutions of a new age; we did not. And they reorganised us.
The colonial period then formalised education as a pipeline for servants of that system: miners, mechanics, clerks, low-level administrators. Again, not education for the frontier, but for someone else’s immediate needs.
Post-independence, the rhetoric changed, but the basic logic remained. We adopted education as a tool to “catch up” – to build industries and institutions that would, in theory, bring us to civilizational parity. Our curricula mirrored foreign models. We focused on producing professionals to fit into an existing global structure.
Sixty years later, we can be honest: the core goals of that model have not been achieved. We remain far from parity in GDP per capita. We are far from parity in knowledge creation per capita. The gaps in research, technology, and capital formation are still large. The development project, as originally conceived, is incomplete.
And yet the world is not waiting for us to finish catching up. The system around us is changing again.
Technologically, we have entered the age of artificial intelligence. The United States and China are pouring hundreds of billions of dollars into building artificial systems designed to be self-improving, self-learning and eventually superhuman in many domains. Estimates suggest roughly 300 billion dollars will be invested in AI this year alone, with that figure expected to double next year. Progress in frontier systems is measured in months, not decades.
This isn’t just chatbots like ChatGPT. It includes embodied AI: robots and automated systems capable of performing almost any physical task a human can do, and at lower cost. One robot doing the work of five or ten people.
Respected institutions in the global North are now openly discussing a future in which the economic benefits of AI are so large that they can fund universal basic income. The idea is that taxes on super-profitable AI platforms will fund a basic livelihood for citizens whose jobs are automated.
That may be plausible in a rich country whose firms own the AI platforms.
But we have to ask: in a world where an AI can do the work of any Zambian lawyer, accountant or engineer, and the AI systems are owned by companies in San Francisco, Seattle or Shenzhen… where does the value go? It goes to shareholders and taxpayers in those jurisdictions.
If a US company owns the AI that outperforms your law firm, the profits – and the tax base for UBI – accrue in the US. What, then, happens to Zambian labour? To our professionals? To our tax base? To our social contracts?
What does education mean in a future where automation is the default, and foreign platforms own the tools?
The honest, uncomfortable answer is that the underlying power dynamics of the world make this scenario very likely. The age of AI is not something we get to vote on. It is already underway.
But inevitability is not the same as destiny. The question we must ask is: what can we do, consciously, in response? Will our future, once again, be defined entirely by foreign powers? Does the world’s frontier have to be our frontier?
This is really a question about our place in the world and whether we choose it, or simply accept it.
Every serious strategy begins with a clear sense of leverage. What is our asymmetric advantage? What do we have – in geography, in resources, in culture, in demographics, in moral authority – that others do not?
If we don’t know our source of asymmetry, we cannot design for it. And if we don’t design for it, others will. They will see it before we do, leverage it before we can, and reap the benefits while we are still debating our five-year plan. That is not because they are evil and we are naive. It is because they are playing to win, and we have often been content simply to play...So we must define our asymmetry clearly.
It doesn’t exist in a vacuum. Take one example. To build AI data centres at scale – the kind needed for frontier models – you need enormous amounts of power, often gigawatts, much of it from natural gas or other firm sources, plus land, cooling, and eventually carbon offsets for the emissions. For all the talk of “digital,” AI is anchored in very physical constraints.
That is an opportunity. In a world obsessed with computation, the value of stable nature, of carbon sinks, of secure land, of reliable sun and wind goes up, not down. Zambia, and Africa more broadly, sit on large endowments of land, sun, water, forests and minerals.
We could simply repeat the old pattern: sell the raw inputs cheaply – power, water, land, carbon – and import the finished systems and decisions. Or we could treat those endowments as the basis for a different social contract: one in which our ecosystems are not just commodities, but leverage for a future in which our children are not disposable.
There will be many such zones of emergent advantage: in energy, in carbon, in biodiversity, in demographics, in culture, even in the freedom to say “no” to certain technologies that are incompatible with our view of what a human being is.
Which brings me back to education.
Educating for the next frontier cannot simply mean more of what we have done for sixty years. It cannot be “let’s produce more lawyers and more accountants and more engineers” without asking what world they are entering and who owns the tools they will be using.
To be educated for the frontier, we will have to abandon our historic naivety.
We will have to teach our children, and ourselves, to ask of every technology and every partnership: is this good in itself, and is it good for us? Does this serve human dignity in our context, or does it quietly erase it? Are we being invited to participate as co-authors of the future, or simply as users, consumers and data sources inside someone else’s paradigm?
We will have to equip them with skills that are aligned with reality, not vibes – with the actual power structures of the world as it is, and the world as it is becoming. That means technical excellence, yes, but also historical literacy, moral courage, and the ability to think in systems and in very long time horizons.
If our ancestors once failed to adapt their skills to the age of exploration and science, we cannot afford to make the same mistake in the age of AI. The frontier will not wait. The question is whether we educate a generation that can meet it on its own terms – or whether we continue to adapt, politely and intelligently, to the plans of others.
The time to be romantic about the past is far gone. We must question the past, determine where we went wrong; course correct—whilst building for the future; and this has to happen fast because no one in the rest of the world will wait.
Postscript:
One of the things that's clear is that the world ahead is one where asymmetric advantages in domains where artificial intelligence doesn't necessarily have a marginal production alpha will make all the difference for places like Africa. The irony is that many of the things that have kept Africa relatively underdeveloped likely become the things that possibly give it advantage. Namely natural resources. Able stewardship of these resources becomes critical. But in that context an atomised approach to resource stewardship likely undermines asymmetry. To that end, OPEC like frameworks become attractive. Of course this entirely ignores the fact that the geopolitical situation is itself also in flux. My essay "the promethan age" speaks to that. Navigating that dynamic and frankly high-stakes environment will take very astute leadership.
