Quick Summary
Not a guarantee but a clear enough view of the direction of travel to act on now. What 2033 looks like, what it means for the people you are building, and why waiting for certainty is itself a decision.
Reading time: 6 minutes.
What would you give to see the future? Not certainty, nobody can offer that. But a clear view of the direction of travel. The ability to anticipate how you should respond before the terrain shifts beneath you.
Based on current trajectories in AI capability, skills displacement, and South Africa’s own labour market data, here is what the world of work looks like in 2033. Informed projections, close enough to act on now.
Jobs Redefined
Jobs no longer have fixed descriptions. The rise of AI agents systems capable of executing entire workflows autonomously, from onboarding employees to tracking supply chains to drafting first-cut financial models has redrawn the map of work entirely. Skills matter more than job titles. Fluid, project-based roles within organisations are the norm. What you can do matters more than where you have been.
AI is a colleague, not a tool handling research, drafting, analysis, scheduling, and first-pass decision-making across most knowledge roles. The human contribution has narrowed to what agents cannot replicate: judgment, context, ethics, relationships, and genuine creative and critical thinking.
The World Economic Forum’s Future of Jobs Report 2025 found that roughly 44% of core worker skills have been disrupted in the previous 24 months. The skills half-life has collapsed.
Source: World Economic Forum, Future of Jobs Report 2025
A qualification tells an employer where you have been. The new territory needs to know something harder to measure: whether you can navigate where nobody has been yet.
The Centaur
The professionals thriving in this environment are not those who resist AI or those who simply follow its outputs. They are the ones who have learned to work with it as a partner directing it, interrogating it, and taking responsibility for what it produces. Researchers call them Centaurs: part human expertise, part AI capability, operating as a single unit.
The concept was pioneered by Garry Kasparov after his 1997 chess match against IBM’s Deep Blue he found that human-AI teams outperformed both humans and machines working alone. Harvard Business School researchers have since applied this framing to knowledge work. In 2033, the Centaur is not exceptional. It is the baseline.
AI-fluent professionals command a 56% wage premium over peers with equivalent experience but without AI skills.
Source: PwC Global AI Jobs Barometer / World Economic Forum, 2026
But Centaurs are not hired. They are built. They develop through years of real work accumulating domain knowledge, building judgment through mistakes, being mentored by people who did the same. The entry-level roles that AI is currently displacing are not just starting points. They are the first stage of the pipeline that produces the Centaurs organisations will need in ten years. Remove that stage and the pipeline does not slow. It stops.
For South Africa, where nearly half of all young people aged 15–34 are already outside employment, education, and training, this is not a future risk. The pipeline that should be building the workforce of 2033 is running dry right now. By the time the demand for Centaurs is undeniable, the people who should have been developing into them will have spent those formative years outside any productive pathway. The fault line is not coming. It is already forming.
The decision in front of every employer
Research modelling South Africa’s BPO sector found that under a proactive scenario where employers invest in reskilling alongside AI adoption the sector could employ over 500,000 people by 2035. Under a reactive scenario, where AI is used primarily to reduce headcount, that number falls by more than 200,000. Same technology. Entirely different outcome. The difference is the decision employers make about whether people are part of the plan.
That decision is not made once. It is made continuously, in every hiring choice, every training budget, every Workplace Skills Plan.
Source: Bridgespan Group / Harambee Youth Employment Accelerator, 2026
An honest word about the limits of any telescope
The projections in this piece may be wrong in their specifics. Things are moving too fast for any forecast to be precise.
Two years ago, every young person asking for career advice was told the same thing: learn to code. It was the surest path. Today AI writes code. Tools like GitHub Copilot have transformed what a coding career looks like in ways that nobody credibly predicted in 2022. Junior developers are among the most disrupted. The profession still exists, but it looks nothing like what the old map showed.
The lesson is not that projections are useless. It is that the specific destination may shift while the direction of travel remains clear. Building people who are adaptive, AI-fluent, and grounded in strong foundations will serve them in 2033 whatever specific form that world takes. Waiting for certainty before acting is not caution. It is falling behind on a trajectory that is already in motion.
The map that got you here will not get you where you need to go. Lift the telescope. Update the chart.
iFundi has been mapping this terrain for 25 years workplace-integrated learning, co-designed with employers, continuously updated, measured by outcomes not hours. Over 20,000 graduates already in the workforce. We do not offer an old map. We offer the intelligence to navigate new territory.
The question
If this is the world your people are heading into, what are you building them for today?
Future-ready is not a slogan. It is a new map. The question is whether you have lifted the telescope in time to see what is on it.
People Also Ask (FAQ)
What is agentic AI and how is it different from regular AI?
AI agents are systems that plan, make decisions, and execute multi-step tasks autonomously managing entire workflows end-to-end rather than just responding to prompts. The shift is from AI as a tool you use to AI as a colleague that works alongside you. That changes what the human in the seat needs to know and be able to do.
What is a Centaur worker?
A professional who has learned to work with AI so effectively that human expertise and machine capability operate as a single unit. The concept was pioneered by Garry Kasparov after his 1997 chess match against IBM’s Deep Blue he found that human-AI teams outperformed both. Today’s research shows the same dynamic in knowledge work: AI-fluent professionals command a 56% wage premium, and the gap is growing.
What does the 2030 jobs data actually show?
The World Economic Forum projects 170 million new roles globally by 2030, against 92 million displaced — a net gain, but only for those with the right skills. In South Africa, Bridgespan Group and Harambee research found that proactive AI adoption in BPO alone could create over 500,000 jobs by 2035. A reactive approach produces less than 300,000. Same technology. The difference is employer intent.
How should South African employers prepare for this shift?
Invest in AI fluency built through real work, not training events. Protect entry-level pathways — they are the first stage of the pipeline that produces the senior talent you will need. And include future-ready skills in your Workplace Skills Plan to access the expanded SETA grant funding under the 2026 SDL reform. The funding exists. The strategic intent is what is missing.
What skills will be in demand in South Africa by 2030?
The evidence points in two directions simultaneously — and both matter. Technical AI skills are becoming non-negotiable across most knowledge roles: AI literacy, the ability to work with and evaluate AI outputs, and for more senior roles, the capacity to design and direct AI agents. These are the skills that command the 56% wage premium PwC’s 2026 research identifies for AI-fluent professionals. But alongside them, the human skills that AI cannot replicate are becoming more valuable, not less. Critical thinking — knowing when the AI is wrong. Creativity — asking the questions the model cannot. Leadership and judgment — taking accountability for decisions that a system can inform but cannot own. As Lesley-Anne Gatter, Global Head of People at Investec, put it: ‘Our people want to learn prompt engineering, but equally they’re asking for empathy, leadership and meaning.’ The
Stefan Lauber
CEO & Founder of iFundi
Stefan Lauber is Co-founder and CEO of iFundi, South Africa’s employer-driven workforce transformation partner, which he built from the ground up over 25 years. He convened South Africa’s inaugural Job Summit and is a contributor to The Heart of Change by Harvard Business School professor John P. Kotter. A former Senior Consultant at Deloitte and researcher at the University of the Witwatersrand, Stefan is also Co-founder of the Foundation Life for All in Switzerland. He has spent his career at the intersection of education, employment, and economic inclusion and writes on the future of work, skills development, and what it will take to build the workforce South Africa needs.