Here's the thing nobody running the teen social media bans wants to say out loud: we are pulling an entire generation off the feed, country by country, and we don't have a reliable way to find out whether it helped. That's not me being contrarian. STAT News asked the question directly in September: bans are passing around the world, so how will we know if they actually improve teen mental health? The researchers' honest answer is a shrug with footnotes.
So that's the core of it, stated plainly. The bans might be a reasonable precaution. They are not a proven treatment. And the gap between those two things matters a lot more than the shouting on either side suggests.
What just happened
Australia went first, passing a law restricting under-16s from social media platforms_Bill). It's the most aggressive version of this idea any country has actually enacted rather than just threatened. Other countries are lining up behind it, and at least one expert told CNBC a U.S. equivalent is "inevitable". That's not hypothetical hand-waving, either: the Kids Off Social Media Act is already sitting in the Senate, waiting for its moment.
The platforms can read a room. Meta rolled out new teen rules of its own, which mostly earned it columns pointing out that the protections are a decade late. Fair.
Meanwhile, the science underneath all of this is nowhere near as settled as the policy momentum implies. The question of what social media actually does to children and young people is, right now, a live academic debate published as a debate, with researchers formally arguing both sides in a peer-reviewed journal. That's about as far from "the science is clear" as science gets. And the latest data on how teens and young adults are actually doing is genuinely mixed, according to the Jed Foundation's recent roundup: some indicators improving, others still worrying. Not a crisis graph pointing straight down. Not a recovery graph pointing straight up. Mixed.
So: the biggest real-world experiment on adolescence and screens ever attempted, launched without a control group or an agreed-on definition of success.
You already ran the experiment
If you're in your twenties, you might notice something uncomfortable while reading this. You were the experiment. You got the feed at twelve, or ten, uncut and unregulated, right through the exact years these laws are now designed to protect. Nobody carded you at the door. Nobody measured you before and after.
And for roughly a decade, the ambient cultural message has been that this broke you. Your attention span, your anxiety, your dating life, your inability to sit through a movie without checking something: all of it filed under "the phones did this." The bans, whatever their merits, carve that message into law. They're a formal declaration that what you grew up inside was too dangerous for children.
Maybe it was! I'm genuinely open to that. But notice what the declaration skips: if the harm were as clean and measurable as the policy confidence suggests, we'd be able to measure the recovery too. We can't, really. That's the whole problem STAT laid out. Teen wellbeing is shaped by a hundred things at once — the economy, the climate, their parents, the news, whether they got cut from the team — and when Australian sixteen-year-olds in 2029 are doing better or worse than today's, every side will claim the credit or dodge the blame, and nobody will be provably wrong.
We've seen this movie at smaller scale. States have been banning phones in schools at record speed, and the most rigorous research found the bans didn't move mental health outcomes much at all. Not because phones are harmless, but because the school day is a small slice of a kid's life. The national bans are that same bet, scaled up, with the same measurement problem attached and higher stakes.
Our take: precaution is fine, certainty is the con
Here's where I actually land, and it's not the fence.
I think trying the bans is defensible. When you suspect something harms kids and the cost of restricting it is moderate, you don't always wait for perfect evidence. We didn't run randomized trials before putting age limits on cigarettes. Precaution is a legitimate mode of policymaking, and anyone sneering that the bans are "anti-science" is holding them to a standard almost no child-safety law has ever met.
What I can't stand is the certainty being sold alongside them. Politicians are not saying "this is a reasonable bet under uncertainty, and we'll honestly evaluate it." They're saying we are saving a generation. That's a claim the evidence doesn't support, from the same playbook that spent your adolescence telling you, with equal confidence, that your screen time had already ruined you. False certainty in one direction fed the panic; false certainty in the other direction will feed the backlash when the mixed data stays mixed. Both leave you, the person actually living with a phone, with worse information about your own life.
And there's a cost nobody budgets for: enforcement means verification, and verification means everyone, at every age, proving they're old enough to scroll. The checkpoint gets built for the kids and then everyone walks through it forever.
What you can actually do with this
Not much about the laws, honestly, unless you enjoy writing to senators. But there's a personal version of this story that you have full control over, and it's the part I'd actually act on.
The reason nobody can prove the bans work is that nobody kept a baseline. There's no clean record of what teen life felt like before, measured honestly, to compare against after. You can run a better study than Australia: sample size of one, but at least the data's good. Notice, specifically, what the feed does to you. Not what it supposedly does to teens in aggregate. Which apps leave you flat and which leave you fine. What a week of heavy scrolling actually feels like versus what you assume it feels like, because those two things diverge more than you'd think.
The easiest way I know to keep that kind of record is to just say it out loud at the end of the day and let something catch it. That's a big part of why we built Joice — three rambling minutes of talking will capture how a day online actually felt in a way a mood-tracker emoji never will. Six months of that and you have something no legislator or op-ed writer has: real evidence about one specific person's relationship with the internet. Yours.
The world just launched a giant experiment on teenagers without a notebook. You don't have to make the same mistake with yourself.