Read Only Internet: Minors Gotta Wait

A Proposal That Goes Beyond Bans or Vapid Vibes

We need to talk, gang. There’s a purported panic—and likely then a burgeoning counter panic—around child safety, censorship, and privacy on the internet. One group, we’ll call them the Haidtists (for reasons that’ll become obvious), think smart phones and the modern, web-based activities are ruining children’s lives. The other, call them the libertarians (definitely not necessarily all card-carrying LP members), are concerned that too much top-down pressure, plus invasive monitoring and user info collection, will be super corrosive for civil liberties.

But let’s back up with some quick scene-setting. Jonathan Haidt has been writing about, and doing advocacy around, the issue of children and tech for years. His book The Anxious Generation quietly started a revolution in how the issue of child safety and what we might call Big Tech (for lack of a better term) are viewed. the emerging Haidtist movement is very well intentioned, but they’ve been—per their critics—far too game to use the levers of government as blunt instruments in their efforts.

What started as a messaging campaign to adjust usage norms and parenting strategies has turned into, by some lights, a crusade against tech companies writ large, especially social media platforms. Some now want to legally kick every kid under 16 off social media entirely, with heavy-handed identity verification policies to boot. Civil liberties, specifically speech and privacy concerns, abound here.

People like those at FIRE and The Argument, including Sarah McLaughlin and Kelsey Piper respectively, think the putative cure on offer is likely far worse than the disease. The former has mostly whiffed on alternative solutions, arguing that the entire age-verification project, root and branch, is a “papers please” privacy catastrophe that’ll be ruinous for speech. The latter proposes we tax pernicious tech practices, feed changes for platforms, and maybe sell special hardware that can’t reach vice sites.

I think all three are, in their own ways, right about something and wrong about something else, which is sort of the whole problem with this discourse. To modify a formulation from Piper herself: the people who admit there’s a real problem mostly propose fixes that are fatally flawed, the people who correctly diagnose the freedom issues mostly don’t acknowledge there’s an underlying problem, and everyone is too concerned with their own frame to see that a true tradeoff will need to have both teeth and limits.

Well, everyone but moi. I don’t think we have to pick a single side of that coin or die. I think the status quo is just poorly minted. We need to cast a new die.

Diagnosis — Everyone’s Half Right

Haidtists have a strong baseline argument. Smart phones are supercomputers in our pockets. Minors arguably don’t need them at all, but if they must have them, they certainly don’t need unlimited access to all the platforms, apps, and sites. But the Haidtists do overstate and get over their skis when they start talking about social media as uniquely or uniformly pernicious. Likewise when they push for policies that sacrifice freedoms online for ALL people just to “protect the children.”

The libertarians recognize this and think the campaign is worth throwing out, the baby and bathwater alike. Some leave the diagnosis conspicuously absent from their analysis, others mention it dismissively and handwave. If this moral panic, as they view it, has produced such excessive and poorly executed policies, surely it can’t be based in any fundamental truths. Surely. That would be far too inconvenient, after all.

Piper, imagining herself the compromiser, frames the whole thing as a quality-control problem: bad content, algorithmically shoveled at kids, making them hooked and miserable. Fix the algorithm, fix the ads, fix the slop, problem solved (ish). I think that’s the wrong disease model entirely. “Slop” implies the fix is better curation. That’s too naive about root causes and too diffident around solutions.

You could hand a 13 year old a perfectly curated, ad-free, chronological feed of nothing but their friends’ vacation photos and they’d still refresh it forty times a day hunting for validation, still get DMed by a creepy stranger, still spiral over who liked what and who didn’t. The rot isn’t in what they’re being shown. It’s in what they’re being invited to do. Post. Wait. Get judged. Get propositioned. Repeat. That loop survives every fix on Piper’s list completely intact.

This is a participation problem, not a content problem. Kids have an absurd surplus of free time and an internet built, largely on purpose, to convert that surplus into engagement metrics. The fix isn’t a better filter on what comes in. It’s cutting off the problem at the source: the users themselves.

Why Minors Should Wait, In a Nutshell

I don’t want to skip over the key arguments for this, especially the behavioral and quality of life factors that affect more than minors themselves. The reality is that underage posters are the biggest single vector, by age, for harassment, abuse, grooming, bullying, and other antisocial behaviors. Not just as the victims, but also as the perpetrators. Isolating them, getting them out of the user-base for most/all of these places would be like kicking a bunch of misbehaving teens out of a restaurant, live theater, or other business they don’t need to be in unsupervised.

A famous tweet goes something like this: everyday I’m subjected to the unwanted behavior or opinions of child on this website is like living through a war crime. Rather dramatic, but I relate to the sentiment. Minors have lower impulse control, disproportionate amounts of free time compared to most regular adults, and put greater risks to themselves and others while active on a given platform. They make the environment worse, and only aging out of bad habits—whilst watching adults model better ones—actually fixes that.

Are adults perfect or blameless here? No. The point is not to put one group on a pedestal, nor to scapegoat another. It’s to say, “let’s try a real change,” and have that change be geared towards positive, prospective participation. Towards watching and waiting. Minors can look, but not touch. They’ll adjust, find better hobbies, stay out of so much internet-based trouble. Meanwhile, adults as users, adults as creators, adults as the drivers of tech companies, will adjust too. Right now, these entities are willfully courting minors, as participants with no perspective or sense, and we see the negative externalities all over. Like giving someone a car or a gun before they’re ready.

Most of the major internet harassment campaigns, some of which led to doxxing and/or suicides, heavily involved minors. Many instances of sexual misconduct, including grooming, transmission of child pornography (even if self-filmed), and other issues started on social media. Minors sexting each other is still illegal, and rightly so. Removing one vehicle for it, especially when that vehicle includes adults they shouldn’t even be talking to, seems beyond prudent. Like driving, drinking, gambling, or even voting, these tech companies and services can wait.

Read Only, No Loopholes

Here’s the pitch: minors get the internet, but as read only. Zero engagement of any kind, full stop. No posts, no likes, no comments, no DMs, no polls, nothing that constitutes participating rather than lurking. If a platform can’t offer a genuine no-account lurker mode (Twitch, many gaming apps, anything that’s architecturally interaction-or-bust), then minors just don’t get that platform. Some places are gonna become adult-only by default, and I’m fine with that. Treat social media participation like gambling, or voting, or signing a contract—it should be a thing you age into, not a birthright you’re owed access to as soon as you can hold a phone.

Crucially, I want the whole internet to start feeling like it’s built for and marketed to adults—the way a bar does, or even a casino. Rather than trying to sanitize every corner of it into a padded room a toddler could wander through safely. That sanitization instinct is the porn-filter problem in miniature, and it doesn’t even work. And pointedly, the attraction factor is half the battle. Once platforms stop being fun to participate in as a minor, because the companies are deliberately avoiding them, plenty of the selection and social effects that got us here start working in reverse. If they can’t monetize minor engagement, they’ll stop designing for minor engagement. The product drifts adult by nature rather than by constant, top-down fiat.

I want to be clear about a few things as well. Firstly, that read-only accounts don’t need to be created per platform. If a place already has logged-out lurking, that’s the read-only mode, no new account infrastructure required; if it doesn’t, thems the breaks. Secondly, as we’ll address further below, these companies would need new TOS or contract modifications and enforcement. Ones where a minor was caught circumventing the policy, or a company was caught looking the other way, faces stiff penalties: IP-bans à la Reddit/Wikipedia for the user, fines and litigation for the business. No need for a nanny state per se, but no more total anarchy either.

Suspected age-violations should be reportable, but the system needs punitive guardrails against brigading, because I am, personally, exactly the kind of high-volume and combative poster who’d be a juicy target for exactly this kind of bad-faith mass-reporting campaign. Something like: reporting privileges are opt-in and earned, the threshold to trigger an actual investigation is a handful of flags (three-ish), a real human staffer reviews before anything happens, and a false or malicious report carries real consequences for the reporter. Flagged accounts stay presumed innocent (unlocked) during review, not presumed guilty. Reddit does the opposite of this with bans and it’s a genuinely atrocious model. We should put our best foot forward here.

On KYC Generally (Where I’m Honestly Mixed)

I’ll admit it plainly: a full-blown KYC (know your customer) verification regime makes my skin crawl. I don’t want that, not for everyone, and McLaughlin’s not wrong that third-party verification vendors get hacked. So here’s where I land. Hard “read only before 18” is the actual norm, non-negotiable. Beyond that, I’m flexible on the mechanism, as long as enforcement has real teeth. No verification apparatus so vast it becomes a national ID system by the back door.

Besides, my actual lever isn’t ID verification at all. It’s account age. Every profile, on every platform, displays creation date, no exceptions. New accounts (say, under 5 days) get max scrutiny. Weeks-old accounts (under 2 months) get a bit of slack. Months-old (under a year) get treated close to fully trusted. This is a behavioral proxy, not an identity check. New accounts that cluster tightly around other new accounts read exactly like a friend group who all signed up together after school, which, funnily enough, is also exactly what a sockpuppet ring or troll farm looks like. Weeding out minors this way weeds out a lot of bad actors as collateral benefit.

Yes, this creates friction for a legitimate burner account too. I’m not gonna pretend otherwise. But it’s friction, not prohibition, same as new accounts already get soft-throttled on several platforms. And to the “but what about whistleblowers” objection: actual institutional whistleblowers should go through the press or legitimate legal channels anyway, not cold-post to socials. Ask ex-QAnon fans how a “government whistleblower” posting on an anonymous new account worked out as a truth-detection method. Yes, there will be tradeoffs that put an onus on good faith actors to essentially earn a good reputation through time and repetition, or just face higher scrutiny for going baby-faced anon. I’m okay with that. Reddit already handles new accounts similarly, and people go through those hoops just fine.

Think of it like a gun waiting period. Nobody’s rights are revoked. You just don’t get full exercise of them the second you walk in the door. Plus, if you’re “angry now” and can’t wait, like Homer on the Simpsons, that’s too freaking bad. Indeed, preventing those people from polluting these places is honestly the system working as intended.

Money Talks, So Let’s Use That

For platforms where actual payment infrastructure is involved (tips, subs, ad accounts, in-app purchases, anything with a card attached), age verification isn’t some new invasive ask, it’s an extension of KYC/AML stuff that already exists for money-transmission reasons. Apple already figured out a version of this for kids racking up App Store bills on mom’s card. Copy that homework. This is the one place I’m comfortable with real ID-adjacent friction, because it’s piggybacking on regulation that already survives legal scrutiny, rather than inventing a new speech-adjacent surveillance regime from scratch.

In general, I’m grateful that I don’t have to whip out an ID everywhere I go. Even when I used my credit or debit cards. But we live in a world where financial crimes, even just kids stealing cards from adults they know (or more egregious cases of fraud with strangers) are running rampant. Platforms that already have cultivated significantly minor-heavy audiences need to get their act together. Clean house, or it’ll be cleaned for them. They can cater to our kids (with guardrails), or they can process our payments in a lax way, but not both. The choices they make will be quite edifying.

Carrots to Go with Sticks

I don’t want this to sound like pure punishment. This policy will need upsides to be ethical, and to get buy-in from stakeholders. So here are a few things…

Liability safe harbor. Platforms that architecturally implement read-only-for-minors get legal cover against child-safety litigation, earned through design instead of paperwork theater. Better defense than “we tried,” which is what every parental control tool currently amounts to.

Compliance as a sales asset. Once this is normalized the way PCI-DSS is for payment processors, “we comply” becomes something platforms can wave at advertisers and payment companies who are already nervous about minor-targeting liability. Get that money, platforms, I don’t care, just get it by doing the right thing for a change.

A family-friendly certification. Industry-run, something parents actually recognize the way they recognize an ESRB rating. Doable via an industry consortium (they all already talk to each other, they’ll figure it out), maybe kickstarted by a private-public or private-NGO partnership to get the first meeting going in a room.

Creators get something too. YouTube already restricts comments on made-for-kids content; extend that logic, but pair it with an actual carrot: algorithmic discovery boosts for creators who build minor-visible, read-only-compatible content, so they’re not just eating an engagement hit for nothing.

Limits, Falsification, and Criticisms

I want to be cognizant of the reality that this entire proposal is about tradeoffs, the allowances we’re willing to make for ethical freedom versus the restrictions and enforcement we’re willing to institute for safety, security, and sanity. I’ll address some of the big ones, as well as the scenarios that would make me rethink this whole plan.

The Telegram Problem (Not Gonna Pretend This Is Solved)

Full honesty time: there’s a real hole here, and it’s called Telegram (also technically Discord, and many other server-based clubhouse networks). These aren’t public, aren’t indexed to search engines, and the companies don’t answer to anybody. Telegram in particular has been rather obnoxious about flouting regulations and defying court orders until the last minute. Principally tho, the issue for us is that we can’t apply read-only rules to something that’s fundamentally a private chatroom. Not without either a massive top-down surveillance apparatus of every server’s contents (which is genuinely authoritarian), or mandatory KYC at account creation for an app that has zero mechanism for policing what happens after you’re in. I’m not bothering with all that. Neither is something I’m comfortable with, for legal and ethical reasons.

So here’s my honest answer, not a shrug: kids restricted off X (formerly Twitter), Twitch, and all the other big public platforms probably do funnel some into Discord-shaped spaces. I’m accepting that trade. Banning minors from bars doesn’t stop house parties, and we don’t respond to house parties by mandating cameras in every living room. Some harm relocates to a smaller venue. That’s the price of not building a surveillance state that follows kids everywhere they might possibly go. I’d also bet, tho I can’t prove it yet, that getting kids off the big, public-facing engagement engines has real knock-on effects even for the private spaces, since a lot of what makes Discord toxic is downstream of drama that started on Twitch or Instagram first.

Discord/Telegram also don’t have the algorithmic attraction engine that Twitch and TikTok do. Nobody’s fed into a Telegram channel the way they’re fed into a For You page. So the migration is to a place that’s inherently less sticky, less discoverable, less optimized-to-hook. Most Discord users go there off of Twitch. If Twitch becomes completely inaccessible, that’s an entire pipeline closed down.

The parental-advisory framing

Some platforms have porn (X, Reddit), some don’t (Substack); some are lawless secret-messaging apps (Telegram), some aren’t. My sincere and pointed message to parents is be advised, use your judgment. Here’s what parents should actually do about the part I can’t fix: the closed-clubhouse tier is exactly where parental awareness still has to do real work, the same way you’d want to know whether your kid’s hangout is a library or an unsupervised basement.

What Would Actually Prove Me Wrong

I don’t want this to be like the worst case that libertarians worry about for Haidtists’ unfalsifiable forever-project. Two things would tell me this framework is failing, and I’m happy to call them out in advance. One: if migration to closed platforms clearly outpaces any measurable drop on the open platforms we’ve restricted, that’s harm relocating, not shrinking, and the whole bet was wrong. Two: if the report-escalation system develops a documented pattern of being weaponized against high-volume adult posters rather than catching actual minors, that’s the system being turned into exactly the harassment tool its critics warned about, and it’d need a serious rebuild.

What I’m betting on, tho, is that making the internet, the various surfaces minors touch the most, a world of plexiglass will actually change behavior for the better. We can’t stop youth from doing dumb or self-destructive things. I get that. But we can put up reasonable rules, and see what happens. Curfews for minors keep them indoors. Age laws for substances create a norm of waiting, even if some try to circumvent this (and risk real consequences, including jail time, in the bargain).

If the two things I’m most concerned about don’t come to pass, I’ll mark this a success. The real challenge is finding the political will to try.

Where This Leaves Us

The disproportionate participation of minors online is bad for them, bad for adults, and only benefits people who want to take advantage of them. The Haidtists want a full ban, with enforcement mechanisms that will harm civil liberties beyond what’s reasonable or arguably even legally sound. This framework tries to address that without going overboard. Minors can look but not touch, like a teen at a supermarket who crosses the liquor aisle or sees tobacco products behind glass at the register.

I’ve put together a solid, if still early-going, look at a plan. Federal normalization is the goal eventually, same trajectory as gambling law or drug law. Realistically, this gets trialed state by state first, the way most durable policy does. And yes, I’m aware that the final version would need to be tight, both procedurally and jurisprudentially. Some sloppily written version of this could get challenged in court. But a posting-and-DM restriction is a very different constitutional animal from a full access ban, or mandatory KYC on every corner; and laws in this space mostly get struck down when they’re written badly or enforced maliciously. That’s not proof the underlying concept is doomed. I will repeat, we just need to try.

So, at the risk of taking a bow, I think I’ve got something here. Not “just parent better,” not “papers please” for the entire internet, nor a fugazi compromise so toothless it needs a trip to a dental professional as well as a lawyer. This is the framework that solves for pernicious incentives, legal or procedural loopholes, and still admits of important concessions. It’s deliberately not meant to please everyone.

A read only internet is my attempt at a third door, one that actually names its levers, admits its costs, and tells you upfront what would prove it wrong. Piper diagnosed the problem and proposed incentive tweaks. Haidt named the stakes and proposed a wall. I’d rather just change what kids are allowed to do than police what they’re allowed to see, and call it a day. Tell me how I did.

---

Sources & Further Reading

The three pieces this responds to

Piper, Kelsey. “Jonathan Haidt Is Wrong About Age-Gating the Internet.” The Argument, July 13, 2026. https://www.theargumentmag.com/p/jonathan-haidt-is-wrong-about-age — The “problem is real, cure is worse” case; her chronological-feed, ad-tax, and vice-blocking-hardware proposals, plus the anonymous-political-speech argument.

McLaughlin, Sarah. “The ‘Papers, Please’ Era of the Internet Will Decimate Your Privacy.” FIRE (The Expression), June 25, 2026. — The privacy-maximalist case; the Australia rollout, the Discord/k-ID vendor breach, the UK “Australia-plus”/VPN material.

Haidt, Jonathan. The Anxious Generation. Penguin Press, 2024. — Foundational text and the four norms (no smartphone before 14, no social media before 16, phone-free schools, more unsupervised play).

Mixed-evidence flank

Electronic Frontier Foundation. “The Science Is Not Settled: How Weak Evidence Is Fueling a National Push to Ban Social Media for Youth.” June 2026. https://www.eff.org/deeplinks/2026/05/science-not-settled-how-weak-evidence-fueling-national-push-ban-social-media-youth — Argues the assumed social-media/mental-health link is weaker than claimed, and that moderate use is closer to a net positive. Also quotes Candice Odgers on the selection effect.

ITIF. “The Flawed Narrative Driving Tech Bans for Kids.” Feb 19, 2026. https://itif.org/publications/2026/02/19/the-flawed-narrative-driving-tech-bans-for-kids/ — Overlapping critique; useful for the “decades of tech moral panics from radio to arcade games” framing.

NCMEC, CyberTipline Data. https://www.missingkids.org/cybertiplinedata — 2024: more than 546,000 online-enticement reports, a 194% increase over roughly 186,000 in 2023, up from 44,000 in 2021. 2025: 1.4 million online-enticement reports, including 800-plus cases where an offender traveled to meet a child, and more than 80,000 concerning sextortion.

NCMEC, “Online Enticement.” https://www.missingkids.org/theissues/onlineenticement — Enticement occurs across every platform type: social media, messaging apps, and gaming.

Madigan, S., et al. “Prevalence of Multiple Forms of Sexting Behavior Among Youth.” JAMA Pediatrics 172, no. 4 (2018): 327–335. https://pubmed.ncbi.nlm.nih.gov/29482215/ — Across 39 studies, 110,380 minors: 14.8% had sent a sext and 27.4% had received one. Adolescents are far likelier to sext romantic partners, past partners, or friends than strangers — native peer behavior, and self-produced CSAM regardless. A 2020 three-level meta-analysis of 79 studies corroborates (14% sending, 31% receiving).

Patchin, J.W. & Hinduja, S. “Summary of Our Cyberbullying Research (2007–2025).” Cyberbullying Research Center. https://cyberbullying.org/summary-of-our-cyberbullying-research — Lifetime cyberbullying rates rose from 33.6% in 2016 to 58.2% in 2025; across 2016–2025, an average of 23.3% of students reported being cyberbullied recently.

Patchin & Hinduja, “2023 Cyberbullying Data.” https://cyberbullying.org/2023-cyberbullying-data — Nationally representative sample of 5,005 students aged 13–17: about 55% reported experiencing cyberbullying in their lifetime, and roughly 27% within the previous 30 days.

Patchin & Hinduja (2022), “Cyberbullying Among Tweens.” Youth & Society. https://journals.sagepub.com/doi/abs/10.1177/02724316211036740 — Among 9–12-year-olds, about one in five had been exposed to cyberbullying as a witness, target, or aggressor.

Thorn & NCMEC, “Trends in Financial Sextortion.” 2024. https://www.thorn.org/research/library/financial-sextortion/ — Analyzed 15M+ CyberTipline reports (2020–2023); sextortion averaged 812 reports per week in the last year studied, and 90% of detected financial-sextortion victims were boys aged 14 to 17.