The World’s First Verifiable Credentials

By Steve Wilson

The field of Verifiable Credentials is one of the hottest topics in the digital identity industry today.  Commonly associated with blockchain and the Self-Sovereign Identity movement, Verifiable Credentials are in fact an old idea.  It is instructive to break down their essential properties and examine the pioneering examples.

Verify What?

A Verifiable Credential (VC) is a technology sub-system for conveying a bundle of information about someone to a receiver, where the information originates from a trusted issuer. Popular examples include mobile driver licences, COVID vaccination certificates, and proof of age, but we have had these sorts of things for many years, including digital passports and chip credit cards.  Even a radio frequecy road toll tag can be considered a verifiable credential of sorts.  

One or two important things are “verifiable” in a moderd verifiable credential.  Firstly, the receiver can verify who issued the credential — because it is digitally signed by the issuer; secondly and ideally, the receiver can verify that the legitimate holder is involved whenever the credential is presented — because of a dynamic digital signature applied by the holder at transaction time.

I stress that VCs are “technology sub-systems” not to make things complicated but because credentials only work as part of a business application within a specific context.  Let’s speak as plainly as possible about this.  The credential is almost always a simple "official" fact about one party of interest to a second party because it authorizes a transaction or business application. For example, If the application is an electronic prescription, then the context is a regulated health system with rules for the types of practitioners who can write various scripts, and the precise forms that the scripts must take. In that context, there are specific credentials that the participants must carry and rules governing how they behave. In every sphere of business there are credentials. 

A Verifiable Credential often digitises a formal certificate of the holder’s bona fides (such as a membership, licence, qualification or official role) and enables the holder to present their credential electronically. It's simplay stated but a little more complex to engineer in the digital domain while preventing fraud, impersonation, credential takeover and so on. 

The first Verifiable Credentials

The earliest cryptographically Verifiable Credentials were probably cell phone SIMs and Chip-and-PIN payment cards.  These microprocessors use public key cryptography to bind a customer to their account details — an International Mobile Subscriber Identity (IMSI) in the case of a SIM, or a Primary Account Number (PAN) for a credit card.  Thus, when the customer interacts with a network, strong proofs are created that tie the customer to their actions, with enough certainty for a rock-solid bill to be generated many weeks later. 

New waves of work are now underway on novel Verifiable Credentials with decentralised key management. The World Wide Web Consortium (W3C) and the Linux Foundation’s Hyperledger Aries are hosting standards efforts; dozens of vendors have launched new VC solutions since 2020.  The common thread in most recent VC development is the use of blockchain and similar crowd-sourced decentralised registers for publishing certain data elements (especially public keys) needed for processing digital signatures.  Yet the basic VC pattern has remained the same for decades: a digitised assertion of a fact about the credential holder, together with a key pair of the holder, is signed by an issuer.

The Main Stakeholders in Verifiable Credentials

The contemporary Verifiable Credentials movement has settled on three principal roles.  

1. The Holder is a first party seeking to access a resource or receive a service based on them being credentialled in a relevant context, by presenting a copy of that credential; in earlier digital identity discourse, the Holder was referred to variously as an end user or subject (note i).
2. The Verifier is a counterparty who will grant the access or provide the service, after satisfying itself that the credential is correct in context, and satisfies any other rules of the Verifier, such as being current and being in the proper hands; historically this role was usually referred to as the Relying Party. 
3. The Issuer who furnishes the credential in question to the proper holders (note ii).

"User" is not a great word for the credential holder, because in many ways, the receiver or verifier "uses" the credential just as much as the holder does! All three stakeholders make use of the technology to safely and reliably digitize their business processes.   

NOTES

1. Some readers find the term “subject” for a participant in an e-business transaction to be politically loaded, even offensive, suggesting subjugation by a more powerful counterparty who might be setting the rules.  However, “subject” is a nice neutral term in grammar, where it means an important part of a sentence construction.  I like the grammatical perspective for its parallels with how authentication relationships are usually formalised or codified.
2. Further, credentials are usually held in wallets. The idea of a wallet has become a broad abstraction for a sub-system that holds the cryptographic essentials behind the Verifiable Credential, but surfaces just the human-meaningful details, such as a credit card number and brand.  The Apple iPhone wallet has become the preeminent example of a digital wallet, but many more specialised examples are seen in government services apps such as MyService New South Wales.

The language in this field has shifted over time, partly in an effort by contemporary digital identity designers to refresh the discourse.  Yet the Holder, Verifier and Issuer are essentially the same three roles as with any formal (rules-bound) business transaction: the first party, the second party and a trusted third party respectively. 

Properties of Verifiable Credentials

For decades it has been widely appreciated by engineers and lawyers that digitisation needs to be done carefully in order to protect the provenance and integrity of the original real-life credentials.  Simply copying a paper certificate or plastic licence card and rendering it as a digital image is clearly risky, because the digital version might be copied, fall into the wrong hands and be falsely claimed, and because digital images are furthermore usually subject to tampering or outright forgery.  The need for copy protection looms large now, with the advent of virtual credit card wallets in mobile phones, digital driver licences, and proof of COVID.  Policy makers, the media and the general public increasingly appreciate that options for digitising real world credentials vary markedly in reliability. The term “verifiable credential” is taken to address these reliability concerns. 

The interesting properties of a cryptographically verifiable credential are as follows:

• it proves the origin or provenance of the credential; that is, the digitised credential names its issuer and moreover is digitally signed by the issuer so that origin is both machine readable and automatically verifiable
• it is tamper-resistant (or tamper-evident) thanks to the digital signature of the issuer
• it is presented peer-to-peer from Holder to Verifier, without any intermediary, and is verifiable on its face by the Verifier, and
• it proves possession (or control over the presentation) by its rightful holder, in cases where the credential private key is access-controlled; classically a private key will be safeguarded in a PIN-controlled smartcard or smart phone, and so receipt of a verified digital signature corresponding to the credential provides strong evidence that the rightful holder was in control of the private key at the time the credential was presented.