The Identity Turnstile: Age-Verification Laws and the Slow Death of the Anonymous Comment

No one hands a driver's license to the clerk at a town council meeting. No one attaches a passport photocopy to a letter to the editor. For most of the internet's short history, the same was true of the comment box, the forum thread, and the anonymous tip line. That convention is now being rewritten, quietly and state by state, in the name of protecting children.

More than a dozen states have enacted some form of age-verification requirement for online content in recent years, and several more are weighing bills this session. The details vary — some target pornography specifically, others reach social media accounts, app stores, or any site that hosts user-generated material — but the mechanics converge on a single demand: prove who you are before you may participate.

How the checkpoint actually works

The enforcement chain has four links, and each one shifts the burden further from the platform and closer to the person.

The weakest is self-attestation, a checkbox that says the user is over eighteen. Legislatures have understandably grown impatient with it. The next link is documentary: upload a government ID, sometimes paired with a live selfie, to a third-party vendor that returns only a yes-or-no verdict. The third is facial age estimation, in which software guesses an age from a photograph and lets the user through without ever confirming a name. The fourth — and the one drawing the most attention from engineers and civil liberties lawyers alike — moves the check to the operating system or the app store, so that every application on a device inherits an age bracket the user cannot easily change.

That last design is the consequential one. A narrow rule aimed at a specific category of content is a burden on that category. A device-level age signal is infrastructure, and infrastructure outlasts the statute that created it. Once the plumbing exists, extending verification from a handful of sites to the general web is an administrative decision rather than a legislative fight.

The chilling arithmetic

Anonymity is not a loophole in the First Amendment; it is one of its recurring subjects. In McIntyre v. Ohio Elections Commission (1995), the Supreme Court struck down a ban on anonymous campaign leaflets and observed that anonymity protects unpopular speakers from retaliation and from what the Court called the tyranny of the majority. Talley v. California (1960) reached a similar conclusion about handbills. Americans for Prosperity Foundation v. Bonta (2021) applied exacting scrutiny to compelled donor disclosure, holding that the state could not demand names merely because names are useful.

The practical stakes are easy to overlook from the vantage of a settled professional life. Consider a nurse who wants to describe understaffing at a hospital, a probation officer with a story about a courthouse, a domestic violence survivor participating in a neighborhood forum, a teenager reading about a subject her parents would rather she not encounter, or a union steward gauging support before a vote. Each of those people can speak today. Whether each can speak next year depends on whether the platform they use accepts a scanned license.

And verification does not merely deter the timid. It redistributes capability. A sophisticated user can route around a checkpoint with a burner device or a paid service; a warehouse worker making a comment during a lunch break cannot. Anonymity becomes a luxury good — available to those with the time, money, and technical fluency to buy it, unavailable to everyone else. That is a strange shape for a right to take.

What the courts have already said

This is not a novel constitutional question, only a novel set of facts. Congress tried to age-gate the internet in 1996 with the Communications Decency Act, and the Supreme Court struck it down in Reno v. ACLU (1997), noting that requiring proof of age would burden adults' access to material the government could not restrict. A successor statute, the Child Online Protection Act, fared no better through a decade of litigation; the courts concluded that age verification was both burdensome and ineffective, and that filtering software was a less restrictive alternative.

The modern statutes differ in technology. Facial estimation and device-level attestation are genuinely new, and courts will weigh them as such. But the doctrinal spine has not moved: a state must show that the means it chose are the least restrictive available, and that the burden it places on adults is proportionate to the benefit it delivers. The 2024 NetChoice decisions, while centered on content moderation rather than age checks, signaled that the Supreme Court takes seriously the First Amendment dimensions of design mandates imposed on platforms. Lower courts are already splitting on how much deference to give the new generation of statutes.

Is there a narrower road?

There may be. Child safety is a compelling interest, and almost no one disputes that minors should be shielded from some material and some product features. The disagreement concerns mechanism, not motive.

A narrower approach might confine verification to the specific functions that pose the specific risk — direct messaging between adults and unknown minors, for instance — rather than to reading or posting generally. It might require data minimization as a condition of the mandate, so that vendors retain nothing and the platform receives nothing more than a binary flag. It might fund independent auditing of vendors, whose failure modes have included leaked identity documents. It might invest in the parental controls that courts in the 2000s identified as the less restrictive alternative.

What it should not do, critics argue, is convert a rule about a category of content into a permanent identity registry. Even a well-run system that discards documents still issues a durable credential — a verified-adult flag tied to an account — and a credential is exactly the kind of anchor that becomes useful for other purposes later.

The adult in the room

The phrase protecting children has become the most reliable legislative solvent in American politics, and its power is not mysterious: the underlying concern is real. But a rule that requires every adult to identify himself in order to read the news, join a local political group, or comment on a school board meeting is not a child-safety measure with incidental effects. It is a change in the default posture of public speech — from open to gated, from unnamed to documented.

Two questions will settle the matter. The first belongs to courts: whether the least restrictive means standard still has teeth when the technology is new. The second belongs to voters: whether they want a country in which speaking requires showing papers. The answer to the second will determine what the first is even asked to review.