We know too well that South Africa’s structural challenges are fundamentally deeper and different to other economies. We are in a seemingly perpetual cycle of despair: a severe, multi-generational unemployment crisis in one of the most unequal and poverty-stricken countries in the world.
Interrogating the value of artificial intelligence (AI) as a cost reducer only is not going to fly for much longer. If we’re to stand any chance at all of chipping away at the triple threats of unemployment, poverty and inequality, business, government and civil society need to be asking how AI can be used to increase productive capacity, create entirely new forms of work and widen access to digital skills – at the very least. Because the decisions we take now will be judged by future generations not by how tightly we managed payrolls, but by whether the seminal technological shift underway through AI meaningfully expands participation and opportunity in our broader digital economy.
A recent global report by the OECD on how people experience new technologies and generative AI is instructive on where South Africa finds itself, and where the opportunities are to integrate generational and educational differences in how technology is being understood and applied.
The data from across 14 countries including South Africa shows that mostly younger, more highly educated people are doing AI-related training which the report attributed to an awareness of career shifts. They found that 46% of professionals aged 26 to 35 undertook formal AI training over the past year, compared to 39% aged 18-25, 38% aged 36-45, and fewer than 20% of workers over the age of 55. Furthermore, they detected that tertiary-educated employees are twice as likely to actively seek out AI learning as those with just a secondary education.
This reveals a growing risk that AI future-proofing is becoming self-selecting, and is being driven more by younger, more highly educated employees, while older, experienced or less educated workers are slipping away.
More than simply noting that older professionals appear less interested in technology than their younger peers, the deeper issue is that South African employers in both the private and public sector at risk of passively allowing AI capability to become voluntary opt-ins.
The same people who already have a baseline of digital confidence, reliable and fast connectivity, spare time, and a natural curiosity are the ones running experiments, testing ideas and introducing innovative solutions. Younger, more digitally fluent, and more highly educated workers are naturally streaking ahead, and the irreplaceable institutional knowledge base of senior leadership is not keeping pace. And the vast majority of our youth, who are not tertiary educated, are also slipping further behind.
This means that AI adoption requires a profound change management re-think and intervention. Older, more experienced workers are more likely to engage with AI when it is explicitly positioned as a mechanism to extend their expertise, reduce administrative drag, and streamline knowledge transfer, rather than as a beacon signal that their experience is suddenly obsolete. And younger, less educated workers need to be included in the prospects that technology and Gen-AI have introduced to the workplace, in ways that matter for their economic hopes and dreams.
But the disconnect goes deeper than just training metrics.
The OECD report also evidences a growing trust deficit. It highlights that over 80% of younger users rate AI positively across usefulness, trustworthiness, and ethical standing. Conversely, older demographics remain overwhelmingly suspicious, hesitant, or responded with a definitive “don't know”.
In practical terms, in a workplace with multiple generations working alongside each other, younger employees are much more enthusiastic to direct deep, cognitive tasks to Generative-AI tools that they implicitly trust, while their managers and senior colleagues view those exact same tools with deep ethical scepticism and a growing mistrust. This can quickly extrapolate to real conflicts between teams: where younger team members may move quickly with Gen-AI generated summaries or data analysis, senior managers question the integrity, confidentiality, reliability or ethics of the output.
The reality is that both sides have legitimate concerns. Blind trust is highly risky, potentially exposing a business to algorithmic bias, data leaks, hallucinations and liability. But sweeping rejection can paralyse productivity and choke innovation.
To counter both the self-selection trap and the trust divide, progressive workplaces must shift their internal learning models. Moving AI training from a voluntary benefit to a highly structured workforce capability programme is a strategic necessity. In practice, this means building role-based learning pathways. A financial analyst, call-centre manager, HR business partner, lawyer, and executive assistant do not have the same Gen-AI training needs. Each person must know exactly how these tools change their work, how they mitigate risk, where they require human judgment calls, and how they enhance their unique value contribution.
Finally, the OECD report indicates that South Africa is among the countries with the highest Gen-AI uptake and most positive perceptions, especially among our youth. This organic youth optimism is a tremendous macroeconomic asset. But optimism alone will not create economic inclusion and without deliberate interventions, Gen-AI can deepen systemic inequality, because the data also shows that those who are most likely to benefit are already educated, connected, and digitally confident.
There are three ways that business and government can harness this enthusiasm and actively counter the risk of creating another layer of digital exclusion:
- Create entry-level AI-enabled roles: Rather than letting AI erase the bottom rungs of the corporate ladder, we must intentionally design junior roles that didn't exist two years ago. Pipelines for roles like data cleaning assistants, prompt library coordinators, customer insight analysts, AI quality reviewers, automation support roles, knowledge-base curators, and digital workflow assistants should be considered
- Augment labour-heavy sectors: True economic multiplier effects happen when we move AI outside head office. There is massive untapped potential in deploying Gen-AI tools to improve productivity, reduce administrative burdens, and more across agriculture, healthcare administration, education support, small business services, logistics, and township enterprise enablement to name just a few.
- Invest in community-based and work-integrated learning: focused and sustainable partnerships with universities, TVET colleges, tech bootcamps and youth employment programmes can continue to be boosted with corporate involvement targeted at making practical AI skills accessible across urban and rural areas.
Right now, true leadership requires an intentional alignment between the promise of technological investments and our broader economic and employment realities. We have abundant opportunities to automate mundane tasks, not human potential. Let’s seize that opportunity, today, for the generations of tomorrow.
