That "Verify Your Age" Button Just Took Way More Than Your Birthday
That "Verify Your Age" Button Just Took Way More Than Your Birthday
This episode is based on our article:
Read the full article →That "Verify Your Age" Button Just Took Way More Than Your Birthday
Full Episode Transcript
Picture a bouncer at the door of a bar. He doesn't need your name. He doesn't need your address. He doesn't need to memorize your license number. He needs to answer one question — are you old enough? — and then he forgets you completely. Now imagine an age check online that works the exact opposite way. It answers that same yes-or-no question, but it quietly keeps your whole identity as the receipt.
If you've ever clicked a button that said "verify
If you've ever clicked a button that said "verify your age," this is about you. Maybe you uploaded a photo. Maybe you snapped a picture of your driver's license. And here's the part that should stop you — you probably assumed that ID is sitting on a server somewhere, with your name on it, forever. That assumption is common. It's also, in the best systems, completely wrong. Today I want to show you the hidden design choice buried inside every one of these checks. So why do two systems that ask the same question end up treating your privacy so differently?
There are really two ways to prove your age online, and they cost you very different amounts of personal data. The first is called age estimation. Artificial intelligence looks at your face — the geometry, the texture, thousands of tiny landmarks — and it guesses how old you are. It's fast. It's low-friction. And when it's built well, it can beat a human's guess. But notice I said "when it's built well." That accuracy falls apart in bad lighting, behind glasses, under makeup, or across different demographics.
Here's a detail that surprised me. According to industry researchers, the people this technology struggles with most are adults aged eighteen to twenty-four. Their faces sit right on the borderline, so the system challenges them and bounces them to a second check. The frustrating irony? That's the exact age group digital platforms want the most. The customers they're chasing hardest are the ones the AI reads worst.
So when the face guess fails, you fall back to the heavy option — age verification. This one asks you to upload a real document. A passport, a driver's license. The system pulls out your birthdate, confirms the document is genuine, and checks that it actually matches you. It's legally airtight. It's also data-hungry. You just handed over your full identity to answer one simple question.
The Bottom Line
And why do so many platforms keep that data around? Because most of them were never built just to check your age. They were built to target ads, or to hunt down fraud. According to the Electronic Frontier Foundation, roughly five to six percent of verification sessions involve someone trying to fake their way through. So the biggest systems fight back by analyzing more than a thousand signals at once — your document, your behavior, your network. The trouble is, all of that data lingers. And lingering data is data that can leak.
So the real privacy question was never "should we check your age." It's whether the system is designed to forget. A privacy-first check answers "over eighteen?" and deletes the proof in seconds. The dangerous one keeps your ID for months — not because it has to, but because nobody told it to let go.
Let me leave you with the simple version. There are two kinds of age checks. One acts like the bouncer — it confirms you're old enough, then forgets you. The other keeps your ID on file long after the question's answered. The Electronic Frontier Foundation put it bluntly this past December — no system today can be fully private, fully accurate, and cover everyone. You get to pick two. Some places, like Washington State, are now writing laws that penalize companies for holding your data too long. So the next time you tap "verify your age," you'll know the real question isn't what you're proving — it's whether they're built to remember. The written version goes deeper — link's below.
Ready for forensic-grade facial comparison?
Full forensic reports with detailed similarity scoring. Results in seconds.
Run My First SearchMore Episodes
UK Digital Identity: 275 Firms Face One New Rulebook
A ninety-nine percent confidence score sounds like near-certainty. But run that same system across a database of a million faces, and it can hand you thousands of wrong answers. The number didn't lie. It just never meant
PodcastDeepfake Detection Companies: 1,200 Traded Faces and Addresses
A group chat with twelve hundred members wasn't just trading fake images. It was trading home addresses. Student IDs. The real names of women who never posted a single photo of themselves online. If you've ever had a fr
PodcastIllinois BIPA: Court Says a Recorded Voice Is Now a Face Scan
A court in Illinois just decided that a company can be on the hook for collecting your voice — even if it never once used that voice to figure out who you are. Not "did they identify you." Just "could they." <break time="
