Even earlier than France permitted laws banning social media for kids beneath 15 final January, 13-year-old Benjamin was already questioning what life with out social media would seem like. “If we need to play soccer, we received’t be capable to set up it. What is going to we do? Ship letters as a substitute?” he joked in an interview for Le Monde.
His response captured the central problem behind the rising wave of youth social media bans: Eradicating entry is one factor; understanding what these platforms imply in kids’s lives is one other.
Inside weeks of Australia’s comparable ban, the nation’s eSafety Commissioner reported that platforms had restricted entry to 4.7 million under-16 accounts. Two months later, although, one in five Australian teenagers beneath 16 was nonetheless utilizing TikTok and Snapchat, based on a parental-control information firm. However even when all kids’s social media accounts had been to vanish, do such bans truly make kids safer on-line?
Governments are moving ahead with out answering that query as they observe Australia’s lead. Indonesia’s child-safety framework, which took effect in March, bars kids beneath 16 from holding accounts on “high-risk” platforms. The U.Ok. authorities has introduced plans to ban social media for under-16s, add default in a single day social media curfews for 16- and 17-year-olds, and prolong child-safety guidelines to cowl dangerous AI options. And on 17 September, the European Fee proposed the EU KIDS Act, which might bar kids beneath 13 from social media, set 15 because the EU-wide minimal age for opening an account independently, and require platforms to point out that their providers are age acceptable and protected by design.
However based mostly on my expertise engaged on Child Online Protection initiatives with the Worldwide Telecommunication Union (ITU) throughout Southeast Asia and the Pacific, I do know the bans don’t deal with the actual issues. As a substitute, we ought to be paying extra consideration to the systems that generate harm within the first place—specifically, recommender algorithms, engagement-maximizing design, opaque moderation, and extractive information practices.
Account removals are usually not the identical as on-line little one security
My expertise engaged on defending kids’s on-line security has taught me three fundamental classes:
First, the general public establishments liable for little one on-line safety usually lack the workers, funds, or technical capability to implement complicated on-line security insurance policies.
Indonesia is illustrative. A 2026 UNICEF evaluation discovered capability constraints amongst service suppliers, long-term funding uncertainty, and a necessity for specialised personnel. On the native stage, some workers lacked digital abilities, whereas funds constraints left some areas reliant on exterior help.
Second, many kids, and infrequently their mother and father, lack the digital literacy and significant pondering abilities wanted to navigate on-line dangers safely. My coverage analysis on little one on-line safety in Indonesia, published earlier this year in Digital Society, discovered substantial gaps that account removals can not restore: Many kids lacked steering on navigating the web safely, and huge numbers didn’t know easy methods to report dangerous experiences.
And third, the platforms have restricted impartial oversight as they establish underage customers, design age-verification methods, and report their very own compliance. In Indonesia, platforms themselves are liable for finishing up age verification, whereas the Ministry of Communication and Digital Affairs oversees compliance. TikTok’s appeals course of for customers flagged as underage, as an example, can require a government-issued ID and selfies, which is an issue as a result of it includes gathering the extra private information on an ID card, past that wanted to verify age. Will authorities regulators be certain that TikTok handles that information responsibly?
The privateness paradox of proving age
All ages-based ban creates an engineering drawback: How can a platform reliably decide {that a} person is sufficiently old, with out intruding on different data? Governments and corporations might use id paperwork, parental authorization, app-store checks, or facial age estimation. Every strategy has trade-offs amongst accuracy, privateness, accessibility, and resistance to circumvention.
There are additionally technical points. One device, facial age estimation, attracts on monumental databases however it’s probabilistic, not actual, as a result of folks range a lot. It’s additionally been proven to misclassify each kids and adults.
The problem shouldn’t merely be to “confirm age.” It ought to be to show that somebody is above a threshold, with out disclosing their id, start date, or different data third events may use to create a advertising profile. The European Commission’s age-verification blueprint challenges firms to confirm ages with out gathering all that extra data.
Privateness-preserving applied sciences provide promising methods to realize this. Zero-Knowledge Proofs (ZKPs) can affirm that somebody meets an age threshold with out revealing their id or actual date of start. W3C Verifiable Credentials are cryptographically verifiable digital claims that may disclose solely the data wanted, similar to “over 16.” And device-based age signals can permit a cellphone or app retailer to share an age vary with out revealing a person’s actual start date. However these strategies nonetheless require rigorous safety testing, frequent requirements, impartial oversight, and clear limits on information retention. In any other case, poorly designed child-safety policies threat creating everlasting id infrastructures wherein companies, not folks, management private information.
The place connection goes when a platform closes
Blocking entry to a platform redirects some younger folks, however not all the time the place anticipated. Early anecdotal stories in Australia pointed to youngsters migrating to smaller, less-regulated platforms like Yope, a sample the Cato Institute flagged as a “whack-a-mole” drawback for regulators. However trade information collected two months later discovered no broad-based shift of that sort, other than a small uptick in WhatsApp use. Many teenagers merely discovered a method to keep on the banned platforms.
This factors to a deeper hole in present society: the erosion of youth “third places“ bodily areas the place younger folks have room to socialize and construct id exterior house and faculty. As these areas have diminished, industrial communications platforms have absorbed that function.
For a lot of youngsters, social media workarounds are merely inconvenient. However for remoted, marginalized, disabled, or LGBTQ+ youth who depend upon on-line communities for help that’s in any other case unavailable, displacement can imply shedding sure sorts of belonging, or having to maneuver to a platform with even weaker oversight.
The best way to design safer on-line methods for kids
If blanket social media bans don’t work, then what is going to? The platforms have created lots of the situations that governments are actually making an attempt to include: engagement-optimized recommenders, intrusive information practices, weak safeguards in opposition to undesirable contact, and options similar to infinite scroll, autoplay, streaks, and chronic notifications.
These design patterns more and more face regulatory scrutiny, together with what’s required beneath the European Union’s Digital Services Act. A 2026 study from the 5Rights Foundation that tracked kids’s gadget use minute by minute discovered that the person interfaces form kids’s consideration, sleep, and well-being in actual time.
A extra sturdy response would regulate these interfaces straight, treating children as legitimate users whose privateness, company, and well-being are required protections, not afterthoughts. Which means designing for security from the outset. One instance can be for kids’s apps to have high-privacy defaults, similar to personal accounts and site sharing switched off for minors. They may even have recommender methods that designate the primary components shaping a feed and provides younger customers extra management over personalization. The European Fee has printed age-appropriate interaction guidelines that restrict unsolicited contact and forestall minors from being added to teams with out consent. Guidelines may additionally prohibit engagement-maximizing features that demand users’ attention, similar to autoplay, infinite scroll, usage streaks, learn receipts, and push notifications, by disabling or limiting them by default.
Governments ought to outline measurable outcomes and fund impartial analysis, platforms ought to give researchers significant information entry, and engineers ought to audit age-assurance methods for bias and information leakage. Colleges, mother and father, and youngsters themselves want a seat in designing the know-how that’s designed to guard kids.
If policymakers nonetheless resolve to take away an infrastructure for youth connection, they need to provide one thing higher in return. Social media bans might scale back some types of publicity to dangerous content material and could also be justified for specific ages, providers, or dangers. However they’re only one device, not a complete substitute for safer design, accountable platforms, digital literacy, institutional capability, and noncommercial digital “third places”—moderated communities, inventive areas, and public-interest platforms designed for youth participation fairly than revenue.
The primary wave of social media restrictions isn’t sufficient to maintain kids protected. Governments are nonetheless measuring what’s best to rely, whereas neglecting harder-to-measure outcomes similar to kids’s entry to protected third locations and meaningful social connection, both online and offline. Till governments can present proof that hurt has truly declined, they’ll maintain mistaking account removing for security.
From Your Website Articles
Associated Articles Across the Net
