Skip to content
Proof of Actor

OUTPUT · ACTOR ATTESTATION

Actor Attestations and Trust Receipts

The output of Proof of Actor is a standardised Actor Attestation: a machine-readable trust receipt describing what could be established about an interaction, and a human-readable Actor Trust Receipt to go with it.

View the API Preview

ACTOR TRUST RECEIPT

Illustrative
Actor
AI Agent
Agent identity
Verified
Principal
Verified organisation
Authority
Verified
Requested action
Inside delegated scope
Human approval
Not required by mandate
Mandate status
Valid
12:44 UTCact_4c7m10...

Actor Attestation

An Actor Attestation is a machine-readable record of the evidence Proof of Actor could establish about a digital interaction: who or what acted, whether a human was present and in control, how much AI participated, and, for agents, who authorised the action and whether it was in scope.

API Preview

A conceptual attestation

One shape for every actor. Fields are illustrative and will change as the design matures with partners.

This schema is illustrative only and is labelled API Preview until the infrastructure is genuinely available.
API PreviewGET /v1/attestations/{id}
// API Preview - schema is illustrative only
{
  "actor": {
    "type": "human",
    "confidence": 0.97
  },
  "personhood": {
    "verified": true
  },
  "identity": {
    "verified": true,
    "disclosure": "private"
  },
  "presence": {
    "continuous": true,
    "confidence": 0.98
  },
  "human_control": {
    "confidence": 0.94
  },
  "ai_participation": {
    "level": "low",
    "confidence": 0.88
  },
  "session_continuity": {
    "confidence": 0.99
  }
}

Design properties

What makes an attestation trustworthy

Claims, not raw data

An attestation states what was established. It is designed not to carry recordings, documents or biometric templates.

Confidence, not verdicts

Every probabilistic claim carries a confidence. Proof of Actor never claims an interaction is 100% human.

Selective disclosure

"Identity verified, private" is a complete answer. The relying party learns only what its policy requires.

Bound to an interaction

Issued for a specific session or action, with a time, so stale evidence is not reused elsewhere.

Verifiable and auditable

Designed to be tamper-evident and retrievable later, so consequential decisions have a reliable record.

Policy-neutral

The attestation describes the actor. The application decides whether that actor is acceptable.

Actor Trust Receipt

Readable by people, not only machines

A receipt makes the attestation legible to reviewers, auditors and the person who performed the interaction. The same evidence, in plain language.

AI participation is graded, never binary

  1. 1No meaningful AI signal observedNothing suggests AI materially shaped the interaction.
  2. 2Possible AI assistanceSome evidence, not enough to state more.
  3. 3Probable AI assistanceEvidence suggests AI materially assisted.
  4. 4Declared AI assistanceThe actor disclosed AI use.
  5. 5Autonomous agent activityAn agent, not a human, performed the interaction.

ACTOR TRUST RECEIPT

Illustrative
Actor
Human
Personhood
Verified
Identity
Verified privately
Presence
Continuous
Human control
High confidence
AI involvement
Low
Session continuity
High confidence
12:41 UTCact_91fk28...

ACTOR TRUST RECEIPT

Illustrative
Actor
AI Agent
Agent identity
Verified
Principal
Verified organisation
Authority
Verified
Requested action
Inside delegated scope
Human approval
Not required by mandate
Mandate status
Valid
12:44 UTCact_4c7m10...

Long-term vision

Agency as visible as transport security

Eventually, digital interactions could carry Actor information much like websites carry transport security today. The objective is not to make AI invisible. It is to make agency explicit.

A user may seeVerified Human
A service may seeACTOR: HUMAN · CONTROL: VERIFIED
An agent transaction may containACTOR: AI_AGENT · PRINCIPAL: VERIFIED · AUTHORITY: VALID · ACTION: WITHIN_SCOPE
An uncertain interaction may showACTOR: UNVERIFIED

FAQ

Frequently asked questions

What is an Actor Attestation?

An Actor Attestation is a machine-readable record describing what Proof of Actor could establish about an interaction: the actor type, personhood, identity disclosure, presence, human control, AI participation, continuity and, for agents, principal and authority, each with a confidence where relevant.

What is an Actor Trust Receipt?

The human-readable representation of an attestation, designed so people such as reviewers, auditors or the user themselves can understand what was established during an interaction.

Is the attestation schema final?

No. The schemas on this site are illustrative and labelled API Preview or Proposed API. The design is evolving with early design partners.

Does an attestation contain personal data?

It is designed to contain claims, not raw data. Selective disclosure is a core principle: an attestation can state "identity verified, private" without carrying the identity itself.

Applications should ask for an Actor Attestation.

Not whether traffic looks like a bot. Help us design the attestation your application would rely on.

Join the Developer PreviewRead the research