Skip to main content

Identity

How can people and organisations prove who they are without surrendering the identity, data, and intelligence built around them?

Identity is sacred—and increasingly easy to imitate. AI can clone a voice, face, or writing style.

Identity, truth, and trust form a single meta-problem: you cannot build consensus if you cannot verify who is in the conversation.

The Bridge

Identity connects the inner game to the outer game. Who you are (character, values, culture) is shaped by community and shapes community in return. Crypto makes this legible — not by replacing identity, but by making it portable, sovereign, and verifiable without surrendering it to a gatekeeper.

LayerWhat it isWhere it lives
InnerCharacter, values, pepeha — who you are before you actAgency
SocialCulture, community, belonging — identity in relationThe Journey
VerifiableDIDs, ZKPs, proof of personhood — identity that travelsThis page

A pepeha declares identity to a room. A DID declares identity to a network. Same function, different scale. The inner identity must be real before the verifiable identity means anything.

The Crisis

Deepfakes create two risks: a specific falsehood may be believed, or authentic evidence may be dismissed as synthetic. Both weaken confidence in digital speech.

What's At RiskThe Threat
Personal likenessVoice, face, writing style cloned without consent
ReputationSynthetic content attributed to real people
Authentication"Seeing is believing" no longer works
Public discourseCannot verify who is actually speaking

The greatest challenge: balancing transparency and privacy while pursuing intents and purposes.

Proof of Personhood

Proof of Personhood (PoP) ensures each participant in a decentralized system is a unique human being. Prevents Sybil attacks where an attacker creates multiple fake identities to manipulate the system.

PropertyWhat It Does
UniquenessEach participant verified as a unique human
Sybil ResistanceMultiple fake identities cannot manipulate the system
EqualityEqual voting power and rewards, independent of economic stake

Methods: Biometric verification (iris scans — WorldCoin), physical verification (pseudonym parties), time-locked wallets, zero-knowledge proofs.

Self-Sovereign Identity

Self-sovereign identity means a person or organisation can control how its identifiers and credentials are used without making one platform the permanent owner. It is an architectural goal, not a guarantee supplied by blockchain alone.

PropertyWhat It Requires
ControlGrant, constrain, and revoke use of credentials
PrivacyReveal only what the decision requires
PortabilityMove credentials without rebuilding identity for each service
RecoverabilityRecover access without handing one provider unilateral power
AccountabilityInspect issuers, verifiers, policies, and revocation paths

Methods: Decentralized Identifiers (DIDs) on-chain, Verifiable Credentials without exposing sensitive data, ZKPs for attribute proof without disclosure.

Sovereign Attention

Not your identity, not your data. Not your model, not your mind.

This is a warning about dependency, not a claim that a platform can own a person. When another system controls your credentials, behavioural history, model objective, and interface, it gains leverage over what you can access and what reaches your attention.

People and organisations need a sovereign layer inside the wider technology stack:

LayerSovereign questionMinimum control
IdentityWho can prove who we are?Hold or govern credentials; choose what a verifier learns
PermissionWho may use them, for which purpose, and until when?Grant bounded access and revoke it
DataWho remembers our behaviour and relationships?Inspect, export, delete, and selectively share records
ModelWhat intelligence learns from that history?Choose the model, context, objective, and retention policy
AttentionWhat gets ranked into focus, and why?Set the intention; inspect and change the ranking rules
ProofDid the system follow those rules?Retain receipts that can be independently checked

The loop changes when these controls move toward the edge:

Platform objective -> algorithm -> your attention

becomes

Your intention -> governed data and model -> your attention

Owning every server is unnecessary. The practical goal is to own or govern the decision layer closest to the person or organisation, while using shared infrastructure under explicit permissions.

Use the six layers as a reusable model. Start at identity and stop at the first layer where the subject cannot inspect, constrain, revoke, or leave. That layer is the current sovereignty boundary.

Zero-Knowledge Proofs

ZKPs prove a statement is true without revealing additional information. The enabling primitive for both PoP and DID.

ApplicationWhat ZKPs Enable
Attribute verificationProve age or nationality without revealing the data
Identity verificationProve credentials without exposing sensitive information
Transaction privacyValidate transactions without revealing details
Access controlProve permissions without disclosing identity
ComplianceMeet regulatory requirements without compromising privacy

A ZKP protects a proof boundary. It does not decide who stores behavioural data, which model learns from it, or which objective ranks a feed. It is one primitive in the sovereign stack, not the whole stack.

Sui's zkLogin is a concrete example: a person can authorise an on-chain transaction with an OpenID credential without publishing the underlying account identifier on-chain. The remaining OAuth, salt, proving-service, application-data, and attention controls still need to be evaluated separately.

Data Sovereignty

Identity sovereignty and data sovereignty are related but different. A person may control a credential while an application still owns the interaction history built around it. An organisation may control customer records while depending on an external model whose retention and optimisation rules it cannot inspect.

Use this five-part test before calling a system sovereign:

  1. Can the subject understand what is being proved or collected?
  2. Can the subject refuse, constrain, export, and revoke it?
  3. Can the subject choose which data and objective direct the model?
  4. Can the subject inspect why information or action was recommended?
  5. Can the subject leave without losing their identity or history?

If any answer is no, name the dependency instead of claiming full sovereignty.

Context

  • Agency — Identity in motion — character plus capability
  • The Journey — Identity is the root of the Theory of Knowledge
  • Culture — Identity is shaped by and shapes culture
  • Pepeha — The deepest prompt: declaring what grounds you
  • The Mycelium — Truth, trust, identity as agent commerce infrastructure
  • Web3 Identity Tech — Implementation stack
  • Zero Knowledge Proofs — The enabling primitive
  • Data — The custody, consent, purpose, and model-context layer
  • Data Footprint — Map where organisational data enters, compounds, and directs agents
  • Sui Technical — zkLogin as a bounded implementation example
  • Verifiable Truth — What identity is built on
  • Identity — The human anchor — who you are before you prove it
  • Truth Recognition — The capability loop that identity anchors
  • Trust — What identity enables
  • Smart Contracts — Deterministic execution layer
  • Principles — Test whether identity infrastructure serves human agency before scaling it

Questions

When AI can clone any voice, face, or writing style, what does "proving you are you" actually mean?

  • If biometric verification (iris scans) solves Sybil resistance but creates a biometric database, has identity become more secure or more fragile?
  • At what point does privacy-preserving identity (ZKPs, selective disclosure) become so frictionless that people stop understanding what they are proving?
  • If decentralized identity works across platforms, who maintains the revocation list — and what happens when they disagree?

Changes my mind: evidence that a person or organisation can preserve equivalent control, portability, privacy, and exit while one application permanently owns the identity record.

Next question: at which layer—identity, permission, data, model, attention, or proof—does your current system first remove the subject's power to leave?