Digital Identity Schemes Are Rolling Out Faster Than the Debate About Them

Digital identity has moved from pilot projects to production. Driving licences, age credentials and residence documents now live in phone wallets in a growing number of places, and the rollout is outpacing the public conversation about how it should work.

The genuinely good idea

The strongest version of digital ID uses selective disclosure. Instead of handing over a document that reveals your name, address, exact date of birth and document number, the wallet answers a single question: is this person over eighteen? The verifier receives a cryptographically signed yes, and nothing else.

Compared with showing a physical card, which leaks everything on it to anyone who glances at it, this is a substantial privacy improvement. It is worth being clear about that, because the technology is often criticised for problems that belong to bad implementations rather than the concept.

Where implementations go wrong

Three failure modes recur. The first is phone-home verification, where checking a credential contacts the issuing authority, allowing that authority to log every time you prove your age or identity. A correct design verifies a signature offline and tells nobody.

The second is over-collection. If a venue can request a full identity document when it only needs an age check, it usually will, because the data is valuable. Preventing this requires rules about what a verifier may request, not just technical capability.

The third is exclusion. Any scheme that becomes the only accepted form of identification disenfranchises people without a modern phone, without reliable connectivity, or without the confidence to use it. Physical alternatives are not a legacy concern, they are a permanent requirement.

The centralisation question

A single national identity system creates a single point of failure and a tempting dataset. Federated designs, where credentials are issued by multiple authorities and verified independently, are harder to build and considerably safer. The choice between them is political rather than technical, and it is usually made without much public input.

What to ask about your own scheme

Does verification work offline? Can a verifier request more than the minimum? Is there an audit log, and who can read it? Can you use a physical document instead without penalty? What happens when you lose the phone?

Those five questions separate a well designed system from a surveillance instrument wearing the same interface.