By Paul Barasa Sachida
In June, I warned that the cyber threat facing African children was outpacing the systems meant to protect them. A week of conversations at the United Nations General Assembly in New York has made that warning feel understated.
Artificial intelligence isn’t going to shape childhood — it already does. Children use it to learn, to play, to talk to each other, to make sense of the world. For a learner in a remote community, a phone can open a classroom and a library that previous generations never had access to.
The same screen opens something else. A child can be sitting in the safest room in the house and still be one tap away from bullying, exploitation, or a stranger who wants something from them. At the Fostering the Future Together meeting, one panelist said something I haven’t been able to shake: we talk about closing the digital divide, but devices and software alone don’t close it — the real gap is community access and knowledge, and until that closes, connectivity just hands children more exposure without more protection. At the Invisible Architecture forum, Romania’s First Lady, Mirabela Grădinaru, made a comparison that stayed with me longer. She likened where we are on child online safety now to where the world was on road safety a century ago, every country building its own rules in isolation, before anyone realized that safe roads only work as a shared, coordinated standard that crosses borders. Online child safety is standing at that same starting point.
I sat in three different rooms that week, the Invisible Architecture forum on safe digital systems for children, Fostering the Future Together’s coalition meeting, and the Coalition for Children’s Rights and Protection conference on AI and child safety. Different audiences, different agendas, but I kept hearing the same thing said in different words: children are still an afterthought in the systems now shaping their lives.
”Invisible architecture” is the phrase from that first forum that has stayed with me. It refers to the decisions no user ever sees, what a platform recommends to a child, what messages it lets through, how it handles a child’s data, whether reporting abuse actually leads anywhere. These aren’t backend details. They are, in practice, the difference between a child who is protected and one who is simply left to figure it out.
Article 3 of the UN Convention on the Rights of the Child says a child’s best interests must be a primary consideration in decisions that affect them. That obligation doesn’t end at the school gate. It has to follow children onto whatever platform, app or AI tool they’re using next.
The scale of this is not abstract. UNICEF’s research across 21 countries found that roughly one in five children had experienced technology-facilitated sexual exploitation or abuse in a single year. Each of those is a specific child who was frightened and didn’t know who to tell, a parent who didn’t see it coming, a teacher who wasn’t sure what to do when they found out.
Africa has more riding on getting this right than most regions, simply because of timing — we’re a young continent and our digital adoption is compounding fast. Wider connectivity and African-built tools are genuinely good things. But handing a child a powerful device isn’t the same as including them in the digital economy; without safeguards, it’s exposure dressed up as opportunity.
None of this is an argument against the technology. AI can support overstretched teachers, make learning accessible to children with disabilities, and flag harmful content faster than any human moderator could. The problem was never innovation — it’s that our safety standards haven’t caught up to what the technology can now do.
Technology companies are the first place this has to change. Safety can’t be a patch applied after a child is already harmed; it has to be part of the product from the first design decision — age-appropriate defaults, real privacy protections, reporting channels that work, and fast action when harm is flagged. And companies operating in African markets need to be far more transparent about how their systems behave here specifically, rather than treating the region as a footnote to decisions made elsewhere.
Governments carry the next piece. Kenya has moved further than many — child-protection legislation, a national response to online child sexual exploitation, and guidelines for safer digital services. But none of that reaches a platform headquartered on another continent. What’s missing is coordination: shared standards across African regulators, and the collective leverage to actually hold global companies accountable.
Schools and parents can’t be left to catch up on their own. Digital literacy needs to be taught with the same seriousness as road safety — because for a lot of children, it’s a comparable risk. Parents need concrete guidance, not general warnings: which settings to check, what the warning signs look like, where to report. Teachers need to feel equipped to act, not just aware that a problem exists. And children need to actually believe that telling an adult will get them help, not blame.
The people most affected are the ones least consulted. Adults keep designing these systems for children without asking children what they actually encounter — the pressure to share a photo, what cyberbullying feels like from the inside, why so many stay quiet. Some of the clearest thinking I heard in New York came from young delegates, not from the panels of adults talking about them.
The measure of UNGA week won’t be how many declarations got signed. It’ll be whether any of it turns into different design choices, enforceable laws, reporting systems that actually respond, and parents and teachers who feel less out of their depth than they do now.
Our children didn’t choose to be born into the age of artificial intelligence. We don’t get to choose that either. But we do get to choose what that age does to them. Africa’s children must not become the world’s unprotected users, or the beta testers of technologies built without safeguards.
They deserve to be able to grow up online the way they’d be allowed to grow up anywhere else — without having to trade away their safety to do it.
Paul Barasa Sachida is a pan African communications and public-policy practitioner. The views expressed here are his own.
Discover more from The Industry
Subscribe to get the latest posts sent to your email.
