Governments around the world are moving toward increasingly interventionist approaches to child online safety. The evidence driving this shift is very strong. Clinicians, researchers and policymakers have documented causal links between social media use by minors and anxiety, depression, disordered body image, eating disorders and substance abuse. According to Africa Newsroom, children online also face sexual exploitation in its many forms: grooming, sextortion and the circulation of child sexual abuse material have proliferated on the very platforms designed to keep young users engaged for as long as possible, with researchers noting that financially motivated extortion rings increasingly target teenage boys aged 14 to 18. Added to this already murky picture is cyberbullying, exposure to violent and pornographic content, and the algorithmic amplification of harmful material. A UN poll of over 170,000 people aged 13 to 24 found that one in three young people surveyed had experienced online bullying, and that one in five had missed school as a result.
This has led to sweeping legislative responses. In November 2024, Australia became the first country to impose an outright ban on social media for under-16s, a law that took effect in December with fines of up to 49.5 million Australian dollars for non-compliant platforms. By June 15 2026, the then UK Prime Minister, Keir Starmer, followed suit, announcing an outright ban on social media for children under the age of 16. The ban is scheduled to commence in the spring of 2027 and will include Snapchat, TikTok, YouTube, Instagram, Facebook and X, but not WhatsApp and Signal. The UK ban pairs the prohibition with restrictions on specific functions, including livestreaming and stranger contact with minors, alongside a £132.5 million programme of offline alternatives for children. Canada, Brazil, Indonesia, France, Spain, Denmark, South Korea and others are either implementing or actively considering their own age-restriction frameworks, according to ABC News.
In the face of governments’ moves to reduce the negative impact of internet access on children, there is one lingering question: whether these frameworks and regulations can actually do what they promise.
This question is especially pressing in Africa, where the risks are no less real, the children no less vulnerable, and the regulatory conditions substantially different.
In Africa, The Risks are the Same, but the Guardrails are Sparse.
Where the UK and Australia are wrestling with how to regulate platforms that are to some extent answerable to domestic courts and regulators, African countries are dealing with a different and perhaps bigger problem: weak enforcement infrastructure coupled with even more acute risks. A 2025 “State of Online Harms” report by the advocacy group Gatefield found that an estimated 68.9 million Nigerians, roughly half the country’s active internet users, regularly encounter cyberbullying, scams, online impersonation, gender-based harassment and child sexual exploitation.
These risks are not evenly distributed, and neither are the policy responses meant to address them. The three African countries that have made the most legislative efforts are Nigeria, South Africa and Kenya, each of which has attempted to build a legal structure for child protection online, with mixed results.
Nigeria’s existing legal structure for online harm is principally the Cybercrimes (Prohibition, Prevention) Act of 2015 (as amended in 2024), but it does not provide a comprehensive framework specifically addressing online harms to children. It criminalises cyberstalking, harassment and impersonation, but researchers examining its application noted that enforcement remains hampered by capacity constraints within law enforcement and the rapidly evolving nature of the harms themselves. South Africa has more child-specific online protections spread across several laws. Its Cybercrimes Act of 2020 similarly criminalises sexual exploitation, grooming and the malicious distribution of intimate images, but according to Legalwise, its rollout has been slow, and victims have historically had to rely on older, ill-fitting common-law remedies such as crimen iniuria or criminal defamation to seek redress for cyberbullying. Kenya has moved the furthest on regulatory architecture: the Communications Authority adopted binding Industry Guidelines for Child Online Protection and Safety in April 2025, covering ICT service providers and aiming to reduce children’s exposure to child sexual abuse material, grooming, cyberbullying, sextortion and radicalisation. The guidelines entered into force in October 2025, and some government officials are reportedly considering restrictions on specific platforms including TikTok and Instagram, though no formal legislative proposals have been tabled.
Lessons from Australia and the UK
Australia was the first to enact the legislation, and as such, its experience in implementing the world’s first social media ban is an instructive stress test . The law took effect in December 2025, and the early results have been telling. By March 2026, Australia’s eSafety Commission found that seven in ten parents reported their child still had an account on a newly restricted platform, according to NBC News. Terry Flew, a professor of digital communication and culture at the University of Sydney, described this as an expected phase of “teething problems” rather than a fatal flaw, arguing that no government was ever going to achieve total disappearance of underage users from every platform on the first day of enforcement.
Critics point out that the Australian bill passed both houses of parliament in nine days, with the public given only a single business day to make submissions, a process the Australian Green Party called “rushed and reckless” and which left consequential terms like “reasonable steps” undefined. Similarly, civil liberties groups have raised their own concerns. These concerns include the following: bans of this kind tend to push youth activity into less visible, less moderated corners of the internet; age verification technology is itself a privacy hazard; and the law was drafted without meaningful consultation of young people, Indigenous communities, or mental health professionals. The libertarian-leaning Cato Institute went even further, warning that the policy will effectively drive tech-savvy teenagers toward VPNs, much as age verification requirements for adult content spiked VPN usage in both Utah and the UK, and that enforcing a ban on VPN usage would require a level of internet policing few democracies would accept. Even academic researchers caution that the underlying science is thinner than the political consensus suggests, arguing that existing studies linking social media use to youth mental illness are too methodologically limited to justify a measure as blunt as a nationwide ban, and that alternative explanations for rising adolescent distress deserve more scrutiny before such sweeping legislation becomes the global template.
Even researchers sympathetic to restriction acknowledge that there are tradeoffs. For instance, Dr Brittany Ferdinands of the University of Sydney has suggested that digital literacy education and parental controls might prove to be a more sustainable solution than a ban, while Dr Christina Anthony, also at Sydney, noted that social media is not incidental to adolescent life but is a part of how many young people build identity and friendship, emphasising that a ban carries real social costs even if it also offers a chance at better emotional regulation. However, one thing that unites both advocates and sceptics is the tacit admission that platforms have failed at self-regulation, thereby prompting governments to respond to the companies’ failure to be particularly responsive to moral suasion.
These are cautionary data points for African policymakers, which are worsened by conditions that make implementation substantially more difficult. If Australia’s eSafety Commission, a well-resourced, dedicated regulatory body, is already struggling to enforce compliance after the law’s commencement, what might that trajectory look like for the underfunded digital regulators across sub-Saharan Africa? If age verification raises serious privacy concerns in Australia, where data protection frameworks are relatively mature, the question of who holds verification data, who has access to it and under what authority becomes considerably more fraught in countries where government surveillance infrastructure is already contested terrain. And if there are concerns in Australia about excessive government intervention online, those concerns sit alongside years of evidence in Nigeria, Kenya, to mention a few, that cybercrime and misinformation laws framed to protect citizens have been deployed to arrest journalists and silence dissent.
There is also a deeper tension that neither the UK nor Australian debates have had to confront in quite the same way. Social media and the internet expose children to very real harm, but access to these spaces also provides information, education, community and avenues for participation and expression that can be difficult to access offline in many African contexts. Where geographic isolation, under-resourced schools, and constrained public information environments are the norm, the digital space can be one of the few places a young person can encounter ideas, opportunities and communities that extend beyond their immediate surroundings. A blanket social media restriction, however well-intentioned, may withdraw access to those goods from children who have the fewest alternatives. The child protection imperative and the children’s rights imperative in African contexts pull in meaningfully different directions. Any well-thought-out regulation has to grapple with both simultaneously.
The AU’s Shaky Response: The Distance Between Architecture and Outcome
The UK and Australia have reached for prohibition, while the African Union has so far reached for harmonisation. In February 2024, the AU Executive Council endorsed the Child Online Safety and Empowerment Policy, which its backers describe as the first region-wide policy of its kind anywhere in the world. The policy does not impose an age-based ban. Instead, it sets out ten policy action areas, spanning institutional capacity, legal and regulatory reform, data protection, business accountability and cross-border cooperation, intended to help member states build national frameworks suited to their own circumstances. It frames the digital environment as inherently cross-border and calls for a whole-of-society approach, noting explicitly that the lack of harmonised regional frameworks has historically been a major stumbling block to coherent child online safety on the continent.
However, the honest assessment is that the AU policy is, for now, an architecture rather than an outcome. It depends on national governments to translate its goals into enforceable law, on regulators that are often underfunded, and on platforms whose African user bases generate comparatively little revenue and, by extension, comparatively little incentive for localised safety investment.
Civil society organisations are working hard to generate granular national data that the AU framework calls for, creating the sort of evidence base that can pressure lawmakers into closing the gap between policy text and implementation capacity. Even then, precedence suggests a future of struggle. The African Charter on the Rights and Welfare of the Child (ACERWC) was adopted in 1990 and entered into force in 1999, yet 36 years later, four AU member states have still not ratified it, and the ACERWC has noted persistent gaps in implementation even among those that have. Health financing commitments made by AU heads of state at Abuja in 2001, pledging 15% of national budgets to health, remain largely unmet across the continent more than two decades later. The pattern, where continental consensus precedes national inaction by years or decades, is one of the more consistent features of AU governance. Therefore, there is no obvious reason or compelling motivation for child online safety to be different unless specific conditions change.
What would those conditions need to be? At minimum, enforceable law at the national level; regulatory bodies with the technical expertise and financial resources to monitor platform compliance; credible sanction mechanisms; and platforms that cooperate rather than stonewall. None of these conditions is currently standard across the continent, and the AU policy, for all its structural sophistication, cannot conjure them into existence through issuing an advisory alone.
What About the Platforms?
Underneath all the complexities around online child safety is a structural asymmetry that should be addressed directly. While African countries are not negligible markets for global platforms, the advertising revenue their users generate remains a fraction of what platforms earn in North America, Europe or even Latin America. As researchers working on content moderation have documented, the consequence is that African markets attract correspondingly less investment in safety infrastructure, local-language moderation, and context-specific enforcement. Major tech platforms like Meta, X and TikTok currently moderate content in only eight African languages across a continent of over 1.5 billion people speaking close to 2,000 languages and dialects. Moderation policies designed in California are applied in African countries. The result is a moderation framework that lacks context-specific policies with inadequate investment in local fact-checking initiatives. For children, this means that the harmful content and exploitative contact they are exposed to may be invisible to the platform’s own safety systems and money is the blindfold.
This is also the strongest argument for the harmonised continental approach the AU is attempting to build, for all its limitations. Individual African regulators negotiating with global platforms will always do so from a position of structural weakness. A country with a population of 20 million and a GDP equal to a fraction of Meta’s quarterly earnings has very limited leverage to compel the company to implement costly safety changes. A continental body representing 55 countries and a combined population of 1.4 billion is a formidable opponent that platforms are more likely to take seriously, particularly as global regulatory pressure on child safety intensifies and the reputational stakes for non-compliance rise.
So, the argument for investing in the AU mechanism is not that it has worked before; rather, it should be that no individual member state has the weight to make platforms respond without it. Because, in the end, whether that pressure produces something closer to the UK’s prohibition model, a less-strict platform accountability regime, or a hybrid built around digital literacy and parental tools, as some Australian researchers now favour over outright bans, will depend more on the enforcement capability of African countries and less on which approach is theoretically superior.
Five Important Points
- As governments move to reduce the negative impact of internet access on children, one lingering question remains: can these frameworks and regulations actually do what they promise?
- Australia and the UK have enacted bans and restrictions on internet use for under-16s, but experts say implementation has not gone according to plan.
- In Africa, Nigeria, Kenya and South Africa have made the most legislative efforts to build child protection online with mixed results.
- The AU’s Child Online Safety and Empowerment Policy largely remains a harmonisation framework still far away from actual adoption and implementation by member states.
- Collective action by African countries is needed to implement stronger policies, drive platform accountability and ensure a much safer internet for Africa’s children.
