When a government bans something people want, the desire does not disappear: it simply finds a quieter route. In this vein, the global rush to ban social media for children has reached the point where South African policymakers are openly asking whether we should follow.
My answer, offered with some discomfort, is that we should study the results before we copy the ambition.
The scoreboard so far
In December 2025, Australia barred under-16s from the major social media platforms. Seven months on, the scoreboard is not flattering.
In June 2026, the Australian government doubled its maximum fine for non-compliant platforms to roughly 99 million Australian dollars, an escalation that indicates how unsuccessfully the first round went.
A peer-reviewed evaluation published in the British Medical Journal found insufficient evidence that the ban had meaningfully reduced young people's social media use. Millions of accounts were closed, yet the children, by and large, stayed connected through virtual private networks, borrowed devices and platforms that happened to fall outside the list.
A government that wishes to place itself between a child and a screen must first persuade a sceptical public that protection is genuinely all it has in mind.
The United Kingdom, which announced its own under-16 ban in June 2026, is now consulting on overnight curfews and feature restrictions, and has asked its regulator to work out what reliable age-checking would even look like. Nobody has solved that problem yet.
We already have tools we barely use
South Africa is not starting from nothing. The Protection of Personal Information Act (POPIA) already treats anyone under 18 as a child and sets strict conditions on processing their information, including the consent of a competent person such as a parent. The Films and Publications Board polices genuinely harmful content.
What we do not have is a law that gates a teenager's access to a platform by age, and before we write one, we should be honest about our capacity to enforce the laws we already have.
Our Information Regulator, the body that oversees POPIA, is young, stretched and although it has issued significant fines, is still building its muscle. Adding a national age-verification mandate to a framework we already under-enforce is a curious way to protect children.
The real question is infrastructure
Here is the concerning part for me. A social media ban is only as real as its enforcement, and enforcing one at national scale requires linking a person's online activity to their verified, real-world identity. That link needs either a government document, a biometric check, or a persistent digital credential.
South Africa happens to be building exactly such a credential right now. Following the President's 2026 State of the Nation commitment, the Department of Home Affairs is constructing a national digital identity system, with the core infrastructure due by March 2027 and verifiable credentials, held in a mobile wallet and secured by biometrics such as a face or fingerprint scan, expected in the 2027 and 2028 financial year.
A children's social media ban would hand that system a ready-made reason to become the everyday key to the internet, and infrastructure built for one purpose rarely stays inside it. Scope creep is inevitable.
A matter of trust
This is, in the end, a question of trust. South Africans have good reason to be cautious about handing the state a master key to their online lives. We have lived through RICA (the Regulation of Interception of Communications Act, which requires every SIM card to be registered and whose surveillance powers the Constitutional Court found wanting), alongside a steady drip of stories about data misused and public systems breached.
A government that wishes to place itself between a child and a screen must first persuade a sceptical public that protection is genuinely all it has in mind.
That is a far harder argument to make in 2026 than it once was, and the burden of proof sits squarely with the state, not with the citizen. The South African state’s ability to regulate, and its propensity for abuse of state resources, is a strong reason for mistrust.
Regulate the design, not the child
None of this is an argument for doing nothing. The evidence that heavy social media use harms young minds is substantial, and it deserves a serious response. The more promising path, though, regulates design rather than identity.
We can require platforms to anonymously identify underage users through content filtering, switch off the infinite scroll and the late-night notifications, and make the algorithmic amplification of distressing content a liability rather than a feature.
We can invest in digital literacy so that children, and their parents, learn to navigate these spaces rather than simply being locked out of them until a sixteenth birthday, at which point they arrive entirely unprepared.
Regulating the product is harder work than banning the user, and it does not photograph as well at a press conference. It is also far more likely to work, and it does not require us to build a surveillance apparatus that we must then trust ourselves never to misuse.
I fiercely oppose nanny-state mentality; but if something is to be done let it be through content management rather than enhancing the state’s control, such as with the Chinese Credit Scoring system.
Before South Africa emulates Australia, we should ask ourselves an honest question: are we protecting our children, or simply building the machinery to watch everyone? I know which of those I would want my own name attached to.
Share
