What AI Literacy Looks Like in a South African Classroom Right Now

 



What AI Literacy Looks Like in a South African Classroom Right Now

Most of what's written about "AI in education" assumes a laptop on every desk and fibre at every home. That's not the classroom most South African children learn in — and any AI literacy programme that ignores this gap isn't actually built for the learners it claims to serve.

Here's what building for the real context actually involves.

The device gap is the starting point, not a footnote

For a large share of South African households, a smartphone is the primary — often only — connected device, not a laptop or desktop. A curriculum written around keyboard-and-mouse interaction, or assuming a device is available at home for practice between sessions, quietly excludes exactly the learners who could benefit most from these skills. Designing for this properly means building around what a child can actually access, not what's convenient to design for.

Connectivity shapes what's realistic to teach

Data costs money, and household data isn't unlimited. A programme that assumes constant, casual internet access for every exercise sets families up to either overspend or opt out. Practical AI literacy in this context leans toward tools and exercises that work in short, deliberate bursts — not ones that expect a child to be online exploring for an hour unsupervised.

The curriculum gap in schools themselves

CAPS (the national curriculum) wasn't written with generative AI in mind, and most schools are working out their own approach in real time, without a shared national standard to lean on yet. That gap is exactly why the extramural/provider model has emerged the way it has — schools bringing in outside expertise for something the core curriculum hasn't caught up to, the same way many did for coding a decade ago.

Township and informal-economy families are not an edge case

A meaningful share of the audience for this kind of programme comes from township and informal-economy contexts — households where digital skills aren't a nice-to-have but a genuine path to opportunity, and where the gap between "knows how to use AI tools" and "doesn't" could plausibly widen an already existing gap. Treating this as the mainstream audience, not a hard-to-reach edge case, changes how a programme should be designed from the outset — not simplified as an afterthought, but built around this reality first.

What "AI literacy" should actually mean here

Not: fluency with the most expensive tools, or assuming constant connectivity. Instead: understanding what these tools are, what they're useful for, their limits, and how to use them meaningfully within real constraints — a smartphone, limited data, a home without a laptop. That's a different, more grounded skill than what most AI-in-education content assumes, and it's the one that actually transfers to the lives these learners are living.

Why this matters beyond one classroom

If AI literacy programmes are designed only for well-resourced schools, the skills gap they're meant to close ends up widening instead — for the exact learners who most need the opportunity. Building for the real South African context isn't a compromise on quality; it's what makes the skill genuinely useful rather than theoretical.

— PatienceR

Comments

Post a Comment

Popular posts from this blog

Why I Built an AI Extramural for a School That Hadn't Opened Yet

Image
  A brand-new primary school here in Gauteng doesn't open its doors until January. But by the time its first learners arrive, it will already have an AI and digital citizenship programme waiting for them — because someone went looking for one before the school even had a finished fees page. That's the short version. Here's how it actually happened. It started with a ChatGPT search, not a Google one Someone on the school's team was putting together their list of extramural providers — the same way any new primary school lines up karate coaches, art teachers, and swimming instructors. But they needed something none of the traditional options cover: AI literacy and digital citizenship for primary-age learners. So they asked ChatGPT who could do it in their area. My name came back. That's the moment that pushed me to start this blog. If AI tools are already the front door people are walking through to find providers like me, I wanted to make sure there was somewher...
Read more

Should My Child Be Using ChatGPT for Homework?

Image
  Should My Child Be Using ChatGPT for Homework? Short answer: probably yes, in some form — the more useful question isn't whether, but how. Used to think, AI tools can genuinely help a child learn. Used to skip thinking, they quietly undo the point of homework altogether. Here's how I'd actually think it through as a parent. The fear is reasonable — but it's aimed at the wrong target Most parent anxiety around ChatGPT and homework boils down to one thing: "Is my child just getting the answer instead of learning?" That's a fair worry. But the tool isn't really the variable — the habit is. A child who copies an answer from a friend, a textbook's back pages, or an AI chatbot is doing the same thing for the same reason: skipping the productive struggle that homework exists to create. The difference with AI is that it's fast, endlessly patient, and available at 9pm when you're not standing over their shoulder. That combination makes bad ha...
Read more

Why Good AI Education Isn't About More Screen Time

Image
  Why Good AI Education Isn't About More Screen Time Hear "AI programme for kids" and your first thought might be another device, another app, another hour of staring at a screen. Good AI education does the opposite — and for the youngest learners, it often involves no screen at all. The assumption worth challenging AI literacy sounds technical, so it's easy to assume it means more time on laptops and tablets. But teaching a child to actually understand AI — what it is, how it works, when to trust it — doesn't require them to be using an AI tool constantly. In many cases, especially for younger learners, it shouldn't. What screen-free AI education actually looks like For foundation-phase learners especially, AI literacy can be entirely story- and play-based: Sorting games that teach pattern recognition — the core idea behind how AI "learns" — using cards, objects, or physical sorting activities Storytelling that introduces ideas like "teachi...
Read more