Baxnet Ideas · Founder note
The Best ID Check May Not Identify You
TL;DR: An age check may need to establish one fact. That does not mean it needs your name, exact birthday, photograph and document number as well.
An online service asks whether you are over 18. Then it asks for a photograph of your passport.
The question has two possible answers. The document offers your full name, exact date of birth, nationality, portrait, document number and expiry date. It is a remarkably generous response to a yes-or-no question.
We have become used to this because identity checks grew up around documents. A passport is a bundle of facts designed for situations where knowing exactly who somebody is matters. We then carried the same bundle into places that may need only one fact from it.
Sometimes the service needs to identify you. Often it only needs to establish that you qualify.
A narrow question deserves a narrow proof
The European Union’s current work on digital identity offers a useful example of a different approach. Its age-verification design is intended to let somebody prove that they are over a threshold such as 18 without disclosing their exact age, identity or other personal details. The Commission says the underlying system became feature-ready in April 2026 and can now be adapted by member states and other implementers ahead of the wider European Digital Identity Wallet rollout.
The plain-English idea is simple. A trusted source has already confirmed something about you. When another service asks a specific question, the wallet can return the relevant answer instead of handing over the source document.
The EU calls this selective disclosure. Its 2026 recommendation on age verification says the default response should be limited to true or false for the age condition being checked, without additional information about the citizen. The wallet rules also require support for disclosing selected attributes rather than every field held in a credential.
That small technical change creates a different way of thinking about identity.
An identity document says: here is a package of facts about me. A selective proof says: here is reliable evidence that the condition you care about has been met.
The distinction travels well. An employer may need to know that somebody holds a qualification, not every result from their education. A landlord may need evidence that an affordability threshold has been met, not a guided tour through months of purchases. A venue may need to know that somebody is old enough to enter without recording their full date of birth.
The answer can be smaller than the evidence used to produce it.
Selective does not mean harmless
There are plenty of situations where a full identity check is legitimate. Opening a regulated financial account, crossing a border and signing certain contracts are not the same as confirming an age threshold.
Even a yes-or-no proof can reveal something sensitive. The fact that somebody completed an age check for a particular service may itself be private. If separate checks can be linked together, a supposedly minimal system can still become a record of where a person went and what they tried to do.
The Commission’s age-verification approach therefore includes protections intended to prevent the proof from becoming a way to identify or track the person. Whether those protections work in real deployments will depend on more than the claim printed on a product page.
A trustworthy design still has to answer ordinary questions. Who requested the proof? What exactly did they receive? Could they ask for less? Can repeated uses be connected? Has the underlying credential expired or been revoked? Can the person inspect a useful record afterwards?
There is another risk too. Once a check becomes quick and convenient, more organisations may decide to ask for one. Data minimisation is not only about shrinking the answer. It is also about resisting unnecessary questions.
Personal intelligence should learn the same restraint
This principle reaches beyond formal identity.
Personal Intelligence products may eventually hold or analyse unusually detailed evidence about a person: financial records, health measurements, work history, conversations and other parts of a private digital life. Their usefulness should not be measured by how much of that material they can pass onwards.
Sometimes another person or service needs one carefully bounded result. A researcher may need one eligible contribution to a benchmark, not the private records behind it. A professional may need a selected report, not permanent access to the archive from which it was produced. The person should remain able to inspect the source and decide whether the result leaves their private space.
Building Mimoto makes this distinction fairly practical. The source material can be years of private conversation, so every result or export raises a simple question: what does the recipient actually need? Mimoto does not need to become an identity wallet for that lesson to matter.
For a surprising number of checks, the complete answer may simply be: yes, the condition is met. You do not need the rest.