“A VPN can change the IP address a verification service sees, but it cannot make an uploaded ID private from that service.”
Depending on the method deployed, an online age check may use your date of birth, a selfie, a government ID, or information from a bank, card issuer, or mobile provider.
What information is retained and for how long is determined by the provider, the website, and the type of data.
For example, a provider might delete your selfie as soon as it’s performed the age verification check, while a website may keep the age result linked to your user account.
This is why a claim like “We delete your photo” isn’t the same thing as promising “We keep no records of age verification checks.” That evidence, any information derived from it, and the outcome are all separate factors and can have separate lifetimes. The same is true for any third-party company that may be involved in the verification process.
This guide focuses on this ambiguity, to help you to tell the difference between, say, a deleted image and a deleted account record.
What personal data can an online age check collect?
| Method | What the check uses | What it can establish |
|---|---|---|
| Selfie age estimation | A facial image and measurements derived from it | An estimated age, age range or whether the person appears to meet a threshold |
| Government ID check | An ID image and details such as date of birth. Some checks also ask for a selfie to match the photo on the document. | Age based on the document, with an identity match where required |
| Credit card check | Card details entered with a payment processor and a validity check with the issuer | Eligibility inferred from the age requirements for that card |
| Mobile network or database check | Depending on the service, a mobile number, name, address or date of birth checked against existing records | Age eligibility or a match against records held by another organisation |
| Open banking age check | Age information a bank holds, accessed with the user’s permission | Whether the person meets the age threshold |
| Reusable proof of age | A digital credential issued after an earlier check | Whether the person meets the website’s age threshold |
What does the website receive after an age check?
For a typical third-party check, the website will ask a specialist to assess your age. Generally, the site visitor is asked to submit evidence to that provider, which will then process it and return a result.
A bank, card issuer, or mobile network may supply information for some age verification methods. Other websites perform their own checks.
For example, last year our own Nate Drake bravely went through the age verification process for the popular adult website XHamster when writing about the UK’s Online Safety Act for hhgttyu.pages.dev.
In this particular case, he was directed to a provider who scanned a photo of his face, then returned an age-related result to XHamster to confirm that he was an adult.
The website in question used a third-party service, so in this scenario the site itself wouldn’t normally receive the selfie - only confirmation of the user’s age.
Of course, that doesn’t mean the website receives no data in these situations, but the amount of information received will depend on the specific method used. For instance, if a site tries to verify your age based on banking information like a credit card, this won’t usually give it access to a full history of all your transactions. That’s why it’s crucial to research the specific method each platform uses to find out what information is requested, which party supplies it, and what is returned.

Results can reveal different amounts of information.
If a result is returned to a website, it could be as simple as a binary yes/no answer to the question of whether a visitor is over a certain age. It may also be an age range, an estimated age, or a date of birth. These categories aren’t interchangeable. For instance, if a platform only needs to verify that you’re over 18, it doesn’t necessarily need to have your full date of birth.
Persona, which is one popular age verification method used by Reddit, is a good case in point. The service’s FAQ states that if you use government ID to verify your age for Reddit, then your date of birth is sent to Reddit. In the case of a selfie, Persona sends an estimated age to them. The provider also claims to erase personal information, including collected images, immediately after the verification process ends in these specific scenarios.
Whichever of these two methods is used, Reddit itself doesn’t automatically receive any photos of users in this case, and can link the verified age and user status to their account. That means you should always check carefully before starting the process to confirm both what the provider sends back and what the website saves after receiving it.
A threshold-only alternative
The European Commission has proposed an age-verification solution that supports much narrower forms of disclosure than providing, say, someone’s full date of birth.
In theory, a visitor to a platform following the EC’s design would first verify their age with an issuer, potentially using methods like government ID, an electronic identity, or a banking app. The age-verification app can then tell a website that the person meets a threshold, such as “18+,” without sending more sensitive information, like their name or birthday.
Naturally, the user still has to disclose certain information but at different stages. In other words, the issuer may need evidence to establish eligibility, but the website doesn’t necessarily need the same evidence.
The Commission has proposed an architecture designed to keep age-proof use from becoming a browsing record for the issuer, but this is a design example, not evidence of how every age verification check works in practice in the EU.
How long do age verification companies keep your data?
This is an excellent question, but there’s no single global retention period for age verification.
That’s why you should always look for the specific rules governing the original evidence, any information derived from it, and the result.
You should also check which company holds each record. For instance, just because you know how long a provider stores selfie images doesn’t necessarily mean that you have any insight into how long the main website keeps your verified-age status.
A good place to start is the method-specific notice shown during the actual check. A company’s identity-verification product, reusable digital ID, and one-off age-estimation service may each follow different rules.
Intermediaries can keep separate records.
Yoti says a mobile-checking partner may keep a record of its request for up to two years. However, this refers to a record of the request, so doesn’t automatically mean that the partner keeps a selfie or government-ID image for two years. This retention period also doesn’t necessarily apply to every age verification method that Yoti uses.
For instance, the mobile method involves more than one website and one verifier. Yoti’s documentation says third-party providers may check the name and date of birth associated with the phone number, for instance.
Before engaging in verification, ask which partner participates in your flow, which fields it receives, and what its request record contains. Different parties may have different retention periods for different types of information.
Exceptions should need specific explanations.
Most platforms that use age verification list exceptions covering manual review, fraud prevention, or legal obligations. You should read through this to identify which policy applies as necessary.
For example, Persona’s blog outlines regulatory requirements and suspicious-activity cases when describing evidence-retention choices. Of course, this is generic advice for businesses, not proof that every website that works with Persona, like Reddit, holds sensitive customer information indefinitely.
The UK Information Commissioner’s Office advises organizations to retain only the personal information needed for the purpose. An over-or-under-threshold result may be enough to confirm someone’s age without retaining the underlying data used to verify it. But this is a data-minimization principle, not proof that every company will store only results and delete them after a fixed period.
If you request that certain data be deleted, make sure to ask about each relevant record rather than assuming it all falls under an umbrella like “my data”. For instance, if you’ve previously submitted a selfie, specify the image, extracted or derived information, verification result, and attempt history. You can also ask about the criteria or retention period for any data that’s being stored.
Your rights and exemptions when it comes to data like this will depend on data protection laws in your jurisdiction, so you can’t safely assume that it will be deleted as soon as you make the request.

What happens to your data if you fail or abandon an age check?
Even if an age verification attempt fails or isn’t completed, a provider or the associated website may still store a record of the attempt.
For example, if you’ve submitted a selfie or a picture of your government ID, then these may follow their own deletion schedules. Closing a browser tab or quitting an app doesn’t prove that a remote server has stopped processing an upload. Nor does not seeing a result, such as confirmation that someone is over 18, prove that no data was recorded.
Yoti’s documentation for developers states that there can be one of several outcomes in situations like these.
Its records can include:
- The session identifier
- Verification method used
- Status
- Error code
- The time the session was created.
Its ‘Query Parameters’ include a ‘status’ field for sessions, though these are designed to help clarify the attempt. They don’t store a copy of sensitive information like selfies. Some valid ‘status’ values include:
- FAIL: A check that’s been completed, but didn’t meet the configured threshold, for the relevant over-or-under checks.
- ERROR: The service couldn’t provide an age result or even calculate whether the threshold has been met.
- EXPIRED: This designates a session that’s no longer usable.
In the case of Yoti, the platform also distinguishes between ending, in-progress, and cancelled sessions. That means if there’s an error that doesn’t automatically mean that a user is underage. If a session expires, that’s also not necessarily proof that you would have failed the age test.
Yoti’s documentation also says that its search interface can retrieve session metrics for up to six months. This doesn’t mean that the platform stores every data category for this period - only specific session data.
Even if a session record exists, that doesn’t establish that an uploaded ID or selfie remains stored alongside it.
By way of example, consider a scenario where you upload a selfie, then close the tab during the age verification process before the result appears.
If you do this, you’ve no way of knowing immediately afterwards if processing finished, failed, or has just remained incomplete. You may be able to find out more by reading the provider’s privacy policy, but if this only refers to completed checks, you may need to contact support to ask when it deletes both the image and the record of the unfinished attempt you’ve made.
Above all, avoid trying to repeatedly undergo the age verification process and submitting documents or selfies before finding out whether this is necessary. Each additional upload creates another submission that you might have to track.
Can your ID photo be exposed in a data breach after an age check?
Yes, if a copy of the image of the ID is stored on the system that’s been compromised. An age appeal can also create a copy of an ID, even when the original automated check only retained it for a short time. A photograph attached to a support ticket may be held in a separate system, where it may be subject to different access and deletion policies.
Discord’s disclosure surrounding a security incident in late 2025 is a concrete example of this.
The breach focused on an unauthorized party that compromised one of Discord’s third-party vendors. Of the accounts impacted around the world, Discord identified 70,000 that may have had government ID photos exposed. The platform also claimed that the vendor had stored those photos for the purpose of reviewing age-related appeals.
Discord practised responsible disclosure by both revoking the provider’s access to its ticketing system and contacting any potentially affected users. Its disclosure also mentioned other potentially sensitive support-related information, including:
- Contact details
- Customer IP addresses
- Messages with support agents
This illustrates why you shouldn’t think of an appeal as the same process as the original verification check. Any data you share will likely enter a different workflow.
If you do need to submit an age-related appeal, find out who processes the ticket, whether the attachment is necessary, and when it will be deleted before attaching any files to your official request via the main website or app.
Don’t attempt to blur out or otherwise hide any details on photos of government ID or other official documents without speaking to the provider first. If a document is rejected, you may have to submit yet another request, potentially increasing the amount of your sensitive data that could be stolen in the case of a breach.
Remember, the goal here is to disclose only what the legitimate process needs, through its approved channel, while understanding for how long any new copy is obtained.
Does a VPN protect the data you submit for age verification?
A VPN can’t hide a selfie, government ID, or other age verification information from a service you choose to share it with.
It also can’t change any service’s retention rules with its own privacy policy, no matter how strongly the VPN’s ‘no-logs’ regulations are worded.
If you’re signed into a website, a VPN also can’t prevent it from linking an age-verification result with your user account if you’ve chosen to go through the process.
A VPN also can’t change how securely or for how long data you submit, like selfies, is stored once you upload it to a website or age verification provider.
Where hhgttyu.pages.dev can help.
These days, all major mainstream age verification platforms, identity providers, and websites implementing age assurance use HTTPS. This means the contents of encrypted web requests containing sensitive information, like biometric data, are already protected in transit.
A VPN can encrypt traffic between your device and the VPN server. This means that third parties like local network observers and your ISP have significantly reduced visibility into destinations carried over the virtual VPN tunnel.
But in situations where you’re uploading data to a mainstream age verification service, the traffic contents are also encrypted by HTTPS, making it extremely difficult for third parties to view the specific data you’ve uploaded by analyzing your network traffic alone.
In a scenario where a hhgttyu.pages.dev user accesses an online verification service, their web traffic is routed through one of our VPN servers. That means the verification service sees the VPN server’s public IP address for that connection instead of your device’s usual public IP.
Unless the service has specific safeguards to block VPN usage, it can normally still process the selfie, ID, or any other information you submit for age verification purposes. The main website can also still attach the age verification to your account, assuming that you’re logged in.
A reliable VPN like hhgttyu.pages.dev can be a powerful privacy tool for protecting your traffic at the network level, but it’s not a substitute for carefully checking what you’re uploading and verifying that the recipient is legitimate.
The verifier and website will still apply their own policies to any records that they hold. A bank or mobile-checking intermediary may also have another policy. None of those policies become the same as hhgttyu.pages.dev’s privacy policy just because the upload passed through one of our VPN servers.
This is something we address in our explanation of our ‘no-logs policy’, as while we don’t permanently retain records of our users’ browsing activity, we can’t change the retention policies of sites, which, for instance, may link an age verification result to a signed-in user account.
Questions to ask before uploading
What am I submitting?
Check what information is necessary to undergo the age verification process. A site may just request a date of birth, ask for a selfie, or even require government-issued ID documents. Just because one option seems more convenient, don’t assume that it will automatically collect less sensitive information.
What will the website receive?
This is especially important for websites that rely on third-party services to verify someone’s age. For example, the provider may see your driver’s license with your name, address, and date of birth, but the website may only see a confirmation that you’re over 18, or even just your age range. Ask whether it is attached to your account or another identifier.
Who else participates?
Read the site’s privacy policy to check for a bank, mobile-checking partner, document-processing provider, or support contractor. Just because one platform has a strong privacy policy that respects users’ data doesn’t necessarily mean that others will follow.
When is each record deleted?
Make sure to separate different categories of data like raw images, derived information, results, and attempt logs. Ask exactly how the system detects when it’s time for them to be erased, and whether unfinished checks or manual reviews can follow a different schedule.
What happens if I appeal?
Check whether support will request another document copy, where the attachment goes, and if its storage policies are different from the age verification process that you already completed.
How can I exercise my rights?
Find the correct contact for the website or provider, as applicable. Save the relevant notice and session reference privately to your device, in case you need to ask a specific question later.
What remains after the check
In this guide, we’ve tried to demonstrate that age checks don’t all try to collect and store the same information.
The original evidence, the result, an intermediary’s request record, and a support attachment can each have a different destination and retention period.
A provider can be truthful with you when it tells you that it instantly deletes your selfie after it’s confirmed your age, while the target website can continue to store the age-related result and link it to your account.
You can choose which process to undergo based on what each party actually receives and keeps, including after a failed attempt or appeal. Repeatedly submitting the same information and documents to providers and websites is often a bad idea: it means there’s more data to keep track of, and potentially places you at greater risk if the provider or site’s servers are breached.
While a VPN can protect your network traffic, it can’t change any information, like selfies, that you’ve already uploaded to websites. Nor does routing your traffic via a VPN server change a provider’s privacy policies to match those of the VPN.
Your rights governing how long data is stored and under what circumstances you can have it deleted can vary based on data protection laws in your jurisdiction. But the question in these cases is not just whether information like a photo disappears, but whether it remains after it does.
FAQ
Yes, if the website offers another method, such as selfie age estimation, a mobile network check or an accepted digital proof of age. Issuing a reusable credential can require a separate initial check.
It depends on how it’s processed. The UK ICO distinguishes between analysing facial features to estimate age and using a face template to uniquely identify someone. Under the UK GDPR, biometric data used for unique identification is special category data.
In the age-check model Ofcom describes, the website receives confirmation of whether you meet the age threshold. Read your bank’s consent screen to see exactly what the intermediary can access.
It can be. An age result linked to an identifiable person or their account can be personal data even if it contains no name or date of birth.
Check the provider’s notice for the specific service. Persona says it uses facial analysis to estimate age and excludes age assurance personal data from model training. Look for an explicit policy statement about training when comparing providers.
In the UK, you can ask the organisation holding the data to erase it. The right applies in certain circumstances and has exceptions. Ask separately about the image, extracted details and any age result the organisation still holds.
We love bringing you this content and hope it helps keep you safe and secure online. Feel free to share it with your friends, too.
Here at hhgttyu.pages.dev we are all about internet freedom, and we are happy to be in a position to bring that to everyone. That is why we give you a 30-day money-back guarantee on our Premium plan. No questions asked and no logs recorded.
If you have any questions, please feel to contact our 24/7 support team either at support@hhgttyu.pages.dev or via live chat.









