Artificial intelligence (AI) has become one of the most talked-about technologies in business, but somewhere along the way, the conversation changed.
By Thabang Chukura, founder and CEO of GenerEd and CEO of CMPNY
The promise of AI was never simply about replacing people. At its best, it has always been about helping people work more effectively by automating repetitive, time-consuming tasks and freeing them to focus on higher-value work. That remains one of the technology’s greatest strengths.
Over the past two years, however, much of the public conversation has shifted. High-profile technology layoffs in Silicon Valley and aggressive investment in AI fuelled a narrative that the technology’s primary purpose was to replace workers rather than support them. Businesses around the world inevitably began asking whether this was the future of work.
I think that’s the fear we’ve imported into South Africa.
Our challenge isn’t deciding whether AI should replace people. It’s deciding how we implement AI in a way that makes our businesses more productive while making better use of the people we already have.
That, to me, is a far more valuable conversation.
In South Africa, where unemployment remains stubbornly high, the idea that success should be measured by employing fewer people has never really made sense to me. Our opportunity isn’t to replace workers. It’s to help the people we already have become dramatically more productive.
That may sound like a small distinction, but I think it changes the entire conversation.
I’ve spent much of my career helping organisations implement technology. Long before GenerEd, through CMPNY, I worked with businesses introducing new enterprise systems and trying to modernise the way they operated. One lesson has stayed with me throughout those projects.
Technology adoption is an emotional exercise long before it becomes a technical one.
People don’t usually resist technology because it’s difficult to use. They resist it when it makes them feel replaceable, or when they’re expected to abandon years of experience without understanding why the new way is better.
I remember one client implementing a policy administration system. The software itself wasn’t the issue. It was a capable system. Adoption suffered because claims agents suddenly had to work in completely different ways from how they’d been trained. The organisation had invested heavily in the technology but hadn’t spent enough time thinking about the people and the processes surrounding it. Eventually, employees found ways around the system. They almost always do.
That experience shapes how I see AI today.
If you’re using AI to replace people whose work you never really bothered to understand, you’re not innovating. You’re outsourcing a management failure to an algorithm.
Before leaders ask what AI can automate, I think they need to ask a far simpler question: where do your best people lose the most time?
In my experience, the answer is rarely the work they’re actually employed to do. We hire people for their judgement, their experience, and their ability to solve difficult problems. Then we fill their days answering the same question for the hundredth time, searching for information buried in a cloud folder, recreating documentation that already exists somewhere else, or waiting for the one colleague who knows the answer.
The skills we pay for sit idle while talented people perform work that requires very little of those skills. That’s the real productivity problem.
Herein lies the real productivity problem. The tragedy isn’t the hours that disappear. It’s talented people spending their careers doing work that never needed their talent in the first place.
Ironically, most organisations already have the knowledge they’re looking for. Years of customer webinars, product demonstrations, onboarding sessions, internal training and project recordings already exist inside the business. Yet employees still struggle to find answers because that knowledge is scattered across platforms, folders and people’s heads. I don’t believe most organisations have a content problem. I think they have a discovery problem.
That’s where I believe AI delivers its greatest value, not by replacing expertise, but by making expertise easier to access. When a new employee can find an answer in seconds instead of interrupting a colleague, that’s valuable. When a support consultant no longer has to answer the same “How do I?” question fifty times a week, that’s valuable. When AI drafts routine documentation or surfaces the right policy without someone spending half an hour searching for it, that’s valuable.
In every one of those examples, the job hasn’t disappeared, the repetitive work has. That’s an important difference. The best AI isn’t the one that attracts the most attention. It’s the one that quietly disappears into the workflow. People stop talking about “the AI” because they’re too busy getting their work done. They find answers faster, they onboard faster, they spend more time solving problems instead of looking for information. That’s what success looks like.
Five years from now, support teams won’t disappear. They’ll become escalation specialists and relationship managers because the repetitive layer of support will be handled elsewhere. Learning and development professionals will spend less time creating endless documentation and more time coaching people. Developers will spend less time writing repetitive code and more time designing better systems. Those jobs won’t become less valuable, they will become more human.
South Africa has an opportunity to approach AI differently from many other markets. We don’t need to copy a narrative built purely around cutting headcount. We can use the same technology to unlock something far more valuable: the judgement, creativity and experience that already exists inside our organisations but is too often buried beneath repetitive work.
To me, that’s always been the real promise of AI, not taking people out of the business but taking the repetition out of work.