Do VCs (Verifiable Credentials) Have No Future?

1.4k VIEWS

A blog post is currently attracting attention in the identity industry. It is a October 15 post by Mr. Hughes, founder of Trinsic, which provided decentralized-ID and Verifiable Credentials (VC) solutions on the Sovrin blockchain: “Why Verifiable Credentials Aren’t Widely Adopted & Why Trinsic Pivoted .” A rough summary follows.

Table of Contents

Why VCs Are Not Widely Adopted and Why Trinsic Pivoted

Overview of Trinsic’s Pivot

IIW38 featured a session titled “SSI Didn’t Work; Trinsic Is Pivoting.”
The audience responded positively, with applause and requests for a livestream.
Industry veterans recognized the session’s importance.

Key Premises

Premise #0—Enthusiastic Support: The author remains committed to the ideals of self-sovereign identity (SSI).
Premise #1—Terminology: Terms such as verifiable credentials (VCs) and self-sovereign identity (SSI) are used almost interchangeably to mean technology that transfers attributes reliably and in a user-centric manner.
Premise #2—Market Reality: Markets are unpredictable; successful products need product-market fit.
Premise #3—Adoption Matters: The best technology is defined by its adoption and impact on users.
Premise #4—Avoid Wasting Time: Innovation is encouraged, while known market failures are highlighted as warnings.

Historical Background

2017 began with the Sovrin Foundation, and Trinsic was founded in 2019.
Initially, hundreds of developers registered, but successful customer cases were limited to 500 out of 1 companies.
Market trends favored proprietary solutions over VCs¹.

Identified Challenges

Interoperability Problems: There is no true interoperability among VCs, and standards are not universally agreed.
User Experience (UX) Concerns: Early VC user experiences were inferior to existing solutions, discouraging adoption.
Fragmented Adoption: The lack of a dominant use case or geographic consistency impedes network effects.

Future Outlook

Government Involvement: Government mandates may promote VC adoption, though only in the medium term.
Timing and Market Readiness: The claim that VCs are premature is challenged by the success of proprietary solutions.
Trinsic’s new strategy focuses not on creating new digital IDs, but on helping companies accept existing ones, anticipating continued fragmentation of the ID environment.

Dick Hardt’s View

The article received much favorable feedback, but some say its identification of challenges is insufficient. One example is this LinkedIn post, which points out the following.

It is unfortunate that SSI, VCs, and decentralized identity were conflated. SSI is fundamentally about giving users control over identity, but the industry focused too much on a technology stack.
When he left Amazon 5 years ago, he examined personal digital identity. Although decentralized identity had already existed for considerable time, he did not expect rapid adoption like OpenID 2.0 or OAuth 2.0 and OpenID Connect (OIDC), because it had the same problems as OAuth 1.0: complexity and failure to leverage existing patterns.
Principal Technical Problems with VCs:
- Users must manage cryptographic keys, creating UX problems and barriers to device migration.
- Issuer information is disclosed; for example, age verification may reveal residence.
- Key reuse makes user behavior trackable. Efforts such as batch issuance seek to prevent this, but make implementation and operation still more cumbersome.
The Biggest Challenge Is the Business Model:
- Users will not pay for wallets.
- Issuers invest in infrastructure, while verifiers obtain the value.
- Existing personal-data sales businesses lack incentives to migrate to VCs.
- Information movement in the backend cannot be made visible.

This broadly matches my view, and I believe the late Vittorio Bertocci, mentioned in Hughes’s article, held the same view.

So What “Proprietary Standards” Were Used?

Another puzzling aspect of Hughes’s article is the absence of OpenID Connect. Itsme, Yoti, ID.me, and Plaid, which supposedly grew by adopting proprietary standards, actually use OpenID Connect. Those are not proprietary standards.

The article also says AI systems exemplified by ChatGPT chose OAuth rather than VCs, but these too use OpenID Connect. If market reality and adoption matter, as Hughes says, OpenID Connect was the superior technology for these use cases.

Do VCs Have No Future?

So do VCs have no future? I am not as pessimistic as Hughes, but his point about fragmented adoption is a problem. OpenID Connect simplifies its technology stack and operations by handling only cases where both issuer and device are online. VCs need a similarly decisive focus, albeit on different areas: use cases outside those where OIDC has proved superior. Today’s VCs try to solve too much. It is like trying to build an amphibious car that also runs on rails.

(Figure 1) An amphibious car that also runs on rails

What I have long wanted from VCs is:

They remain usable if the issuer goes offline or ceases to exist.
They remain usable when the device is offline.
Users are not confused about which wallet to use.
Unless subsidized with public funds, value flows back from verifiers to issuers.

The goal is to focus on use cases meeting these requirements and establish a technology stack and ecosystem.

In addition:

Issuer information should be disclosed only as group information.

That would be even better. Perhaps issuers using group signatures could solve this? A question for everyone at the Sako Laboratory.

Focusing here first may reveal a path to success.

Also, as I have long argued, query languages must be radically simplified. Presentation Exchange is far too complex. Even OpenID claims syntax was too complex to gain adoption. I think DCQL is also too complex. Initially, only its simplified subset—still as complex as Claims Syntax—would be enough.

I still have much more to say, but let us stop here for today.

One more thing: I do not know whether it relates to Trinsic’s departure, but the Sovrin blockchain has also been scheduled to shut down. It is a good example showing that blockchains do not guarantee permanence.

See you next time!

Footnotes

called proprietary, though many are based on OpenID Connect