There is something almost everyone agrees on: children spending hours scrolling through social media are not doing their mental health any favours. What nobody can quite agree on, however, is what to do about it and whether the proposed solutions might end up being more damaging than the problem they are meant to solve.
Across Western countries, rates of anxiety, depression, self-harm, and suicide among adolescents particularly girls began climbing sharply around 2010 to 2012, a period that coincided almost exactly with the mass adoption of smartphones and social media platforms. The correlation is difficult to ignore. The causation, however, remains fiercely contested. Researchers and policymakers continue to debate whether the mental health deterioration seen in young people can be attributed specifically to platforms like Instagram and TikTok, or whether other forces the lingering psychological toll of pandemic lockdowns, declining family stability, economic precarity deserve equal blame.
Still, ask most Canadians with years of personal experience on social media whether prolonged use has ever made them feel worse about themselves, and the honest answer tends to come quickly. So, the pressure to act is real. The question is how.
The most closely watched attempt to answer that question is currently playing out in Australia. On December 10, 2025, the country became one of the first in the world to enforce a sweeping social media ban for users under 16. Major platforms Facebook, Instagram, Snapchat, TikTok, X, YouTube, Reddit, Twitch, and Kick among them are now legally required to take “reasonable steps” to prevent underage users from creating or maintaining accounts. Platforms found to be in breach of the law face fines of up to AU$49.5 million.
In the weeks following the ban’s introduction, approximately 4.7 million suspected underage accounts were removed. On paper, that sounds significant.
The reality on the ground told a different story.
Surveys conducted in early 2026 found that more than 60 percent of under-16s who had accounts before the ban had already found their way back onto at least one restricted platform. The methods were not sophisticated borrowed phones, a parent’s ID, a false date of birth entered into an unverified age declaration field, a VPN that rerouted their location, or in some cases, a printed mesh mask designed specifically to confuse facial recognition scans.
Notably, Australian law explicitly prohibits platforms from requiring government-issued identification or a government-accredited digital ID to verify age. To comply, platforms have instead turned to a patchwork of alternatives: behavioural analysis, device signals, and facial age estimation technology. The result is a ban that, within weeks of its introduction, was already being widely circumvented.
This brings the debate to its most uncomfortable point.
If age verification is not robust, the ban is largely symbolic. Millions of underage users return through the back door while the legislation sits on the books, offering the appearance of child protection without the substance. But if the verification is robust if it is made airtight enough to actually keep teenagers out then an entirely different set of problems emerges, ones that do not affect only minors.
To meaningfully enforce an age-based social media ban, one of a small number of technical approaches must be employed. A government-mediated login system would require users to authenticate their identity through a centralized portal each time they access a platform. This would hand governments an extraordinary volume of metadata who is logging into what, how frequently, from where that most Canadians would find deeply uncomfortable were it attached to their bank records or medical history.
The alternative is to place the verification burden on the platforms themselves. This could mean periodic facial scans matched against submitted photo ID, or continuous behavioural surveillance monitoring typing patterns, touch pressure, content consumption habits, and other signals that machine learning systems can now reliably use to estimate a user’s age. Researchers have documented statistical correlations between typing speed, error frequency, and age. Biometric systems can already estimate age from skin texture, facial bone structure, eye shape, and hairline position.
What this means in practice is that Canadians would be surrendering some of the most sensitive and permanent data that exists their faces, their fingerprints, their behavioural signatures to corporations headquartered in foreign jurisdictions. Many of those corporations are subject to laws that can compel them to share user data with their own governments. Unlike a compromised password, biometric data cannot be changed. Once it is leaked, sold, or demanded by a foreign intelligence apparatus, the exposure is permanent.
The legal dimension complicates matters further.
Canada’s Charter of Rights and Freedoms protect freedom of expression under section 2(b), and Canadian courts have recognized that this protection extends beyond the right to speak it includes the right to receive information. For many young Canadians, social media platforms are now among the primary channels through which they encounter news, participate in public discourse, access educational content, and engage with viewpoints different from their own. A mandatory ban that restricts this access raises genuine constitutional questions that no government has yet answered persuasively.
One does not need to be a civil liberties absolutist to take that concern seriously.
Advocates for the ban often present the choice as binary: either legislate minors off social media, or do nothing. But critics argue that Canada has not yet made a serious investment in less invasive strategies that evidence suggests can work.
Aggressive public campaigns discouraging early smartphone adoption, phone-free school environments through at least Grade 9 or 10, and improved parental monitoring tools have each shown measurable impact on youth mental health in multiple studies. Holding platforms legally liable for the addictive design features they deploy specifically to keep young users engaged infinite scroll, algorithmically optimized content feeds, notification triggers is another avenue that remains largely unexplored in Canadian law.
Parents, after all, bear responsibility for their children’s digital lives regardless of what legislation exists. Even under the most comprehensive possible ban, a determined teenager with a willing parent or an indifferent one can find a workaround. Australia has already demonstrated that.
Canada now finds itself at a crossroads that many democracies will face in the years ahead. The instinct to shield children from documented harm is not only understandable it is admirable. But the mechanisms required to enforce that shield at scale carry their own serious costs, costs that fall on adults and children alike, and that do not disappear once the legislation is passed.
An ineffective ban solves little while creating false assurance. A genuinely effective one, however, may require the construction of a digital identity infrastructure that most Canadians would reject if it were proposed for any other purpose.
The cure, as the debate increasingly makes clear, must not be worse than the disease.
