UK vs EU Digital ID: An Architecture Scorecard

1. Two roads to the same destination

The UK and the EU are both building national-scale digital identity, but from different starting points and with different politics. The EU's scheme is the EU Digital Identity Wallet, mandated by the eIDAS 2.0 regulation: every member state must offer it to every citizen by the end of 2026, with large-scale pilots already running across 26 member states and 350-plus organisations. The UK's scheme — announced September 2025, once branded the 'BritCard' — has had a far rockier birth, lurching from 'mandatory for right to work' to 'voluntary' inside four months, with a consultation that closed on 5 May 2026.

Marketing aside, an identity system lives or dies on a handful of architectural properties. This scorecard grades both on those properties rather than on press releases. The questions are the same ones we tell readers to ask of any verification system: can the issuer see where you used it, can your uses be linked, what is retained, and who ultimately holds the keys.

2. The five criteria that actually matter

1. Unlinkability. Can two separate uses of the credential be correlated into a profile? This is the single most important privacy property. A system with unlinkable presentations cannot be turned into a movement tracker; one without it can.

2. Selective disclosure. Can you reveal one attribute ('over 18', 'resident here') without exposing the rest of the document? Real selective disclosure minimises what each verifier learns.

3. Issuer visibility. Does the issuer get pinged at the moment of each use (so it can log your activity), or can the wallet present credentials offline with no phone-home?

4. Retention and minimisation. What data is stored, by whom, and for how long? The privacy-preserving design keeps a yes/no on the verifier side and nothing centrally.

5. Jurisdiction and mandate trajectory. What legal regime governs the data, and is the system genuinely voluntary or on a path to compulsory? Legal architecture is as decisive as technical architecture.

3. The scorecard

The headline: the EU's design names the right privacy properties as goals, but delivery is fragmented across 26 member states. The UK's design is more centralised by default — GOV.UK One Login concentrates identity in one account — which makes the privacy outcome depend heavily on engineering choices that are not yet public.

4. Where the EU is genuinely ahead — and where it isn't

The EU deserves credit for writing unlinkability and selective disclosure into the regulation itself, and for funding the cryptographic research to back it — including 2026 work combining trusted execution environments with zero-knowledge proofs for self-sovereign credential use. Cross-border interoperability was demonstrated in December 2025, which is a real engineering achievement.

But 'the regulation says so' is not the same as 'every wallet does so.' With 26 member states and hundreds of relying parties, the privacy floor will be set by the weakest implementation that still claims compliance. The risk is a patchwork in which some wallets are genuinely unlinkable and others quietly log. The age-verification mini-wallet, shipping this summer, is the first real-world test of whether the privacy promises survive contact with deployment.

5. The UK's specific risk: centralisation and the 'voluntary' ratchet

The UK's architecture risk is structural. GOV.UK One Login is a centralised identity broker, and centralisation is the enemy of unlinkability: when one account mediates access to many services, that account becomes a natural choke point for logging and correlation. Nothing about a central broker requires surveillance, but it makes the surveilling design the easy default and the privacy-preserving design the harder, more expensive exception.

The political risk is the 'voluntary' ratchet. The sequence — announce mandatory, retreat to voluntary under pressure, retain a 2029 right-to-work mandate in the small print — is exactly how a nominally optional system becomes effectively compulsory. When a credential is required to earn a living, 'voluntary' is a technicality. The consultation that closed in May is the public's main lever; the design decisions it informs will set the privacy baseline for a generation of British identity infrastructure.

6. What this means for you, today

Neither system is finished, and that is precisely why scrutiny matters now. If you are in the EU, push for wallets that demonstrably implement unlinkable, offline presentation — and treat the summer age-verification app as the canary. If you are in the UK, the design phase and the 2029 mandate are still live; the unlinkability and centralisation questions are the ones to put to your MP.

In the meantime, identity hygiene and metadata hygiene are within your control. A verifiable, audited VPN keeps your network metadata out of the correlation pool while these systems mature — and the provider that scores highest in our evidence matrix on jurisdiction and audited no-logs is also one we partner with: Proton VPN (Swiss, independently audited, post-quantum). We disclose the relationship; the ranking is formula-driven from graded evidence and does not move for commission. Identity systems are being built right now, in public, with the privacy properties still up for grabs. The cryptography to do this well exists. Whether it gets used is, for a little while longer, still a choice we can influence.

7. References