A digital identity can make a public service feel almost invisible: prove an attribute, sign a form or open an account without repeating an in-person document check. The same infrastructure can also connect activities that were previously separate, exclude people whose records do not match and turn a convenient credential into a de facto requirement. The engineering question is therefore inseparable from a civil-liberties question: who can learn what, under which authority, and what happens when the system fails?

Governments are moving from isolated login programs toward reusable wallets and federated credentials. In July 2025, the US National Institute of Standards and Technology finalized revision 4 of its Digital Identity Guidelines after two public drafts and nearly 6,000 individual comments. The revision added controls for injection attacks and forged media, incorporated synced passkeys and subscriber-controlled wallets, and strengthened fraud, privacy and customer-experience guidance in NIST's revision overview.

The European Union is pursuing a common wallet. Regulation 2024/1183 requires each member state to provide at least one European Digital Identity Wallet under common specifications, with the EU's legal summary setting an end-of-2026 objective. The wallet is intended to store and present identification data and verified attributes across borders, including with selective disclosure according to EUR-Lex.

Proving less can provide more privacy

Traditional online identity often copies a physical document, revealing far more than a transaction needs. An age check may expose a name, address, birth date and document number when the relying party only needs to know that a person is over a threshold. Attribute credentials can disclose the answer rather than the underlying record. Zero-knowledge proofs can go further by demonstrating that a statement is true without revealing the data used to establish it.

The EU regulation explicitly calls for privacy-preserving techniques, including zero-knowledge proof, and requires selective disclosure. It also says wallet providers should be unable to observe transaction details, must keep wallet data logically separate from other service data, and must provide a dashboard showing relying parties and data exchanged in the regulation. These are architectural limits on surveillance, not just promises in a privacy notice.

Data minimization also improves security. A verifier that never receives a full birth date cannot leak it. A system that uses pairwise identifiers can make routine correlation harder. But cryptography does not decide whether an attribute request is necessary. Policy must constrain relying parties, set retention periods and penalize coercive or excessive collection.

Inclusion is part of assurance

Higher assurance can reduce impersonation while making access harder for people with incomplete records, limited connectivity, disabilities or no compatible device. Biometric matching may add convenience but produces false rejections and creates sensitive data that cannot be reissued like a password. Remote proofing can be attacked with synthetic media; in-person proofing can impose travel, time and documentation burdens.

The World Bank's identification principles pair robustness with universal access and "protecting user privacy and control through system design". They also call for technology neutrality, independent oversight and grievance mechanisms. That combination is crucial: a technically accurate system can still be unjust if an error blocks benefits and there is no fast human appeal.

The EU framework therefore prohibits directly or indirectly limiting access to public or private services for people who choose not to use the wallet and requires appropriate alternatives in Regulation 2024/1183. The practical test will be whether those alternatives remain affordable and usable. A paper or in-person route that exists only formally does not preserve meaningful choice.

Digital Identity Systems Must Balance Convenience and Civil Liberties — supporting image
Digital Identity Systems Must Balance Convenience and Civil Liberties — supporting image

Federation changes where power sits

Identity systems divide roles among credential issuers, wallet providers and relying parties. Federation can spare every service from maintaining a sensitive proofing database and let users reuse credentials. It can also concentrate power in a small number of identity providers whose outage, policy change or compromise affects many services.

NIST's 2025 core guidance treats assurance as a risk decision across identity proofing, authentication and federation rather than a single universal strength in SP 800-63-4. A low-risk information service should not demand the same identity evidence as a high-value financial transaction. Proportionality limits both friction and data accumulation.

Open standards and interoperable credentials can reduce vendor lock-in, but governance must follow the data. Issuers need audit rules; relying parties need enforceable purpose limits; wallets need independent security testing and rapid revocation; users need transaction histories and redress. Public code can support scrutiny, yet an open-source application does not reveal how a production service logs, correlates or shares information.

Trust requires enforceable boundaries

The United Nations' 2024 Universal Digital Public Infrastructure Safeguards Framework was developed as actionable guidance for safe and inclusive systems, alongside the Global Digital Compact adopted by 193 member states and territories according to UNDP. Its significance is that identity cannot be assessed only as a software procurement.

Before scaling a credential, governments should publish its legal purpose, threat model, data flows, exclusion testing and remedy process. They should measure false rejection, abandonment and appeals, not only successful authentications. Independent authorities need power to investigate misuse, and optional uses must not quietly become mandatory through market pressure.

Convenience and liberty are not inherently opposed. Selective disclosure, decentralized storage and narrow assurance can improve both. The conflict arises when a system optimizes only the successful transaction and ignores the information it exposes, the power it centralizes and the person it leaves outside.

Sources