My Phone Asked for ID. My Laptop Didn’t: What Age Verification Is Really Checking

The Reddit post opened normally on my laptop, but the same link stopped at “Confirm your age” on my phone. I was signed into the same account, sitting at the same café table, and well past the age the platform was trying to establish.

My first assumption was simple: the laptop knew my age and the phone did not.

That explanation lasted until I signed out on both devices. When I reopened the link in a fresh laptop session, the age gate appeared there too.

The laptop had not decided I was an adult. It had simply been carrying a different combination of stored session data and network signals.

That distinction explains why age verification can appear on one device but not another. The platform may be looking at the account, but it is not relying on the account alone. It can also consider what the app or browser remembers, what age information the operating system makes available and which country the connection appears to come from.

Two devices belonging to the same person can therefore receive different treatment without either device being faulty.

Why This Started Happening More Often

For UK users, these mismatches became much more noticeable after stronger age-assurance duties took effect on 25 July 2025. Services covered by the Online Safety Act were required to introduce highly effective age checks for pornography and certain other material considered harmful to children. By June 2026, all ten of the UK’s most frequently visited pornography services had introduced checks, while some other services had chosen to block UK visitors instead.

The effect was not confined to dedicated adult sites. Platforms carrying mixed user-generated content also had to decide when particular posts, communities or media required an age decision.

That turned a legal obligation into a very ordinary technical frustration: a page that opened yesterday might ask for identification today, and the result could differ between an app, a mobile browser and a laptop.

VPN use rose sharply around the start of the UK requirements. Ofcom recorded daily usage climbing from roughly 650,000 people before the July deadline to more than 1. million in mid-August. Many of those people were not trying to become anonymous in an abstract sense. They were trying to make a page behave as it had before.

My own mismatch came from three practical differences.

The laptop browser had an older session. The phone app had a fresh one.

The laptop was using café Wi-Fi routed through a work connection that exited outside the UK. The phone was using a UK mobile network.

And the app and browser did not share all the same local information.

Once those differences were visible, the apparently irrational result began to make sense.

The Platform Was Comparing Signals, Not Screens

Reddit says its age-assurance process may use information such as account age, an entered birthdate, previous verification, email-related signals, activity on the service and age information available through app stores or external providers.

Apple’s age-assurance framework adds another possible difference. On supported iPhones, an app can request an age range or learn whether a user is eligible for an age-restricted feature. A desktop browser does not automatically receive the same app-level result.

Meanwhile, a browser may remember an earlier choice through cookies or similar local storage. The UK Information Commissioner’s Office explains that these technologies allow a website to recognise a device and retain information about earlier activity. An app can maintain a separate session even when it uses the same account.

The connection itself can also change the decision. A laptop on Wi-Fi and a phone on mobile data normally have different public IP addresses. Those addresses are commonly used to estimate a visitor’s country, although IP geolocation is not the same as a precise physical location.

I could not see the platform’s internal filtering rules, so I could not identify the exact weight assigned to every signal. But the practical evidence was clear: when the session and apparent network region changed, the age-gate result changed with them.

Public support discussions show the same kind of friction without adding much mystery to it.

The useful question was therefore no longer, “Why does my phone think I am younger than my laptop?”

It was, “Which signal is different, and can I make the two devices present a consistent one?”

The First Fix Worked on Only Half the Problem

I began with the least disruptive option: a free browser VPN extension on the laptop.

It seemed reasonable. The problem was appearing on a web page, the extension installed quickly, and I did not need to create another paid account.

After connecting through a non-UK route, I reloaded the page. It opened.

For a moment, that looked like the solution. Then I picked up the phone.

The Reddit app still showed the age prompt.

The extension had changed the laptop browser’s route, not the phone’s connection. It also did nothing for other apps or browsers on the laptop. When the extension later switched routes, the site’s behaviour changed again.

That result was more useful than it first appeared. It showed that changing the apparent region could remove the immediate gate, but a browser-only tool could not keep two devices consistent.

I did not need a longer list of server countries. I needed one route that covered the whole phone, could be repeated on the laptop and would remain connected while the network changed underneath it.

The Route That Made Both Devices Agree

I installed OnlydogVPN at that point.

The app did not begin by presenting a wall of flags and asking me to guess which server might reproduce the laptop result. I chose a situation-based preset intended for restricted browsing and connected without creating a conventional email-and-password account.

Then I reopened the same post on the phone.

The prompt was gone.

The thread loaded, including the image that had previously been hidden. I scrolled through it, opened another post in the same community and returned to the original link. The result remained consistent.

That observable result mattered more than the protocol description. Still, the design behind it explained why this attempt felt different from the browser extension. The connection used HTTP/3-based transport with additional obfuscation, while the preset removed the need to work through server and protocol combinations manually.

The service was not proving my age. It was removing the regional signal that had triggered the gate in that browsing session.

The next problem was the laptop. I wanted both devices on the same type of route, but I did not want to create another login, recover a password or repeat the subscription process while preparing to leave the café.

The app displayed a verification code for adding the second device. I entered it on the laptop, selected the same preset and opened the page again.

This time, the phone and laptop produced the same result.

That was the point at which the problem felt solved. Not because every age-assurance system had been defeated, but because the two devices I was actually using had stopped contradicting each other.

A smaller benefit became visible as I packed up. The phone moved from café Wi-Fi to mobile data. The page paused briefly, then continued loading without dropping back to the ordinary UK route. I did not need to open the VPN app, reconnect and discover halfway through the thread that the gate had returned.

The recovery mattered because mobile browsing rarely happens on one perfectly stable network. Phones move between home Wi-Fi, public hotspots and cellular data. A route that works only until the next network change does not truly solve a device-consistency problem.

What the Test Did—and Did Not—Change

There are reasons some users will still prefer a larger provider. This service has fewer locations, a shorter public history and fewer independent reviews than the most established VPN brands. Someone who requires a particular country, extensive published auditing or a large support operation may place more weight on those factors.

A major full-device VPN could also produce the same practical result when it provides a suitable route on both devices.

But the browser extension had already shown why the familiar comparison points were not enough in this situation. It offered an impressive list of locations and successfully opened one page, yet it solved the problem on only one screen.

The smaller app solved the problem I actually had: the phone and laptop needed to present a consistent network region without separate browser configurations, repeated account setup or another round of server guessing.

Age verification appeared on one device because the platform was not asking one portable question called “Is this person an adult?” It was making a new access decision from the evidence available in that particular app, browser, session and connection.

Once I understood that, matching the devices’ settings mattered less than making their decisive network signal consistent.

For this problem, a stable full-device route across both screens mattered more than proving that both screens belonged to the same adult.