# Proof of Actor

> Proof of Actor is a proposed trust infrastructure for determining whether a digital interaction is being performed by a human, a human using AI, or an autonomous AI agent, and for establishing who authorised the action.

Proof of Actor is in development (Developer Preview). Nothing on this site describes a generally available product: API and SDK examples are labelled API Preview or Proposed API and are illustrative.

Key facts:

- Category: Actor Trust Infrastructure (also described as Human and AI Agent Trust Infrastructure).
- Core question: Who or what is actually acting?
- Actor states: Human; Human + AI; Authorised Agent; Unverified Actor. No state is universally better: trust is contextual and the relying application sets the policy.
- Trust stack: personhood, identity, presence, actor continuity, human control, AI participation, delegation, scope of authority, intent, accountability.
- Verification primitives: Proof of Human (is a real human here?), Proof of Control (is that human controlling this interaction?), Proof of Authority (if an AI agent is acting, who authorised it and what may it do?).
- Output: the Actor Attestation (machine-readable) and the Actor Trust Receipt (human-readable).
- Principle: Proof of Actor supplies trust evidence; the customer determines acceptable policy.
- Privacy principle: Prove what matters without revealing what doesn't.
- Core belief: AI does not need to disappear from digital interactions. It needs to become attributable.
- Not: bot detection, a CAPTCHA replacement, KYC, deepfake detection or AI-text detection. Those may become signals inside the platform; none describes the problem.
- Outputs are confidence and evidence signals, never absolute verdicts.

## Start here

- [Proof of Actor](https://www.proofofactor.com): Trust infrastructure for a world of humans and AI agents: know who or what is acting, whether a human is present and in control, and who authorised the action.
- [How Proof of Actor works](https://www.proofofactor.com/how-it-works): The Actor model, the four Actor states, the ten-layer trust stack, contextual policy and the Actor Attestation that an application receives.

## Verification primitives

- [Proof of Human](https://www.proofofactor.com/proof-of-human): The human-verification primitive: evidence that a real human is behind an interaction, with identity optional and privacy-preserving.
- [Proof of Control](https://www.proofofactor.com/proof-of-control): The human-control primitive: evidence that a present human is actually controlling an interaction, continuously, and how much AI participates.
- [Proof of Authority](https://www.proofofactor.com/proof-of-authority): The agent-authorisation primitive: which AI agent is acting, which human or organisation it represents, what it may do and whether this action is inside that mandate.
- [Actor Attestations and Trust Receipts](https://www.proofofactor.com/actor-attestations): The machine-readable Actor Attestation and the human-readable Actor Trust Receipt that describe what could be established about an interaction.

## Developers

- [Developers (API Preview)](https://www.proofofactor.com/developers): The proposed Proof of Actor API and SDK design: verifications, Actor status, agent authority, explicit human intent and attestations. Developer Preview, design evolving.

## Use cases

- [Use cases](https://www.proofofactor.com/use-cases): Where knowing who or what is acting matters: hiring, financial services, agentic commerce, marketplaces, dating, social networks, reviews and video calls.
- [Hiring and remote assessments](https://www.proofofactor.com/use-cases/hiring): Establishing that the expected candidate personally performed an assessment or interview, with session continuity and transparent AI-assistance signals.
- [Financial services](https://www.proofofactor.com/use-cases/financial-services): Requiring verified human control or explicit human intent before consequential transactions and high-risk account actions.
- [Agentic commerce](https://www.proofofactor.com/use-cases/agentic-commerce): Verifying which human or organisation stands behind a shopping or procurement agent and whether a requested purchase is inside its mandate.
- [Marketplaces](https://www.proofofactor.com/use-cases/marketplaces): Distinguishing genuine users, AI-assisted users and authorised agents on two-sided platforms without blocking legitimate automation.
- [Dating](https://www.proofofactor.com/use-cases/dating): Evidence that the person behind a profile and a conversation is a present human, without exposing identity documents to other members.
- [Social networks](https://www.proofofactor.com/use-cases/social-networks): Letting platforms label accounts and posts as Human, Human + AI or AI Agent, so agency is explicit rather than guessed.
- [Reviews](https://www.proofofactor.com/use-cases/reviews): Future review signals such as Verified Human Review and Verified Buyer + Verified Human, in a world of AI-generated reviews.
- [Video calls and deepfake risk](https://www.proofofactor.com/use-cases/video-calls): Establishing that the person on a video call is the expected participant, present and in control, regardless of how the media was produced.

## Company

- [Research](https://www.proofofactor.com/research): Open research questions, the competitive landscape, standards Proof of Actor builds alongside, and essays on Actor Trust.
- [About Proof of Actor](https://www.proofofactor.com/about): Mission, vision, manifesto and principles: AI does not need to disappear from digital interactions. It needs to become attributable.
- [Security](https://www.proofofactor.com/security): How Proof of Actor approaches security during development, what it publishes and keeps private, and how to report a vulnerability.
- [Privacy by design](https://www.proofofactor.com/privacy): Prove what matters without revealing what doesn't: selective, minimal, privacy-preserving attestations.
- [Early access](https://www.proofofactor.com/early-access): Join the Developer Preview or apply to become a design partner.

## Essays

- [A Trust Model for Humans and AI Sharing the Internet](https://www.proofofactor.com/blog/a-trust-model-for-humans-and-ai-sharing-the-internet): A complete trust model for AI agents and humans online: Actor states, a ten-layer trust stack, attestations and policy for human and AI agent trust at scale.
- [What Is Agentic Commerce, and Why It Needs a Trust Layer](https://www.proofofactor.com/blog/agentic-commerce-needs-a-trust-layer): What is agentic commerce? A clear definition, the trust questions merchants face with AI shopping agents, and how delegated authority and explicit intent help.
- [Authentication Is No Longer Enough: Authentication vs Authorization vs Actor](https://www.proofofactor.com/blog/authentication-is-no-longer-enough): Authentication vs authorization explained clearly, with a comparison table, plus the third question modern services now face: who or what is actually acting.
- [From Authentication to Actor Attestation](https://www.proofofactor.com/blog/from-authentication-to-actor-attestation): A technical guide to actor attestation: claims, confidence, freshness, selective disclosure, agent delegation and the trust receipt a relying party consumes.
- [Human, Human + AI, Agent: The Internet Needs a New Identity Model](https://www.proofofactor.com/blog/human-human-ai-agent-a-new-identity-model): Non-human identity covers service accounts and API keys. AI agent identity needs more: a model that separates humans, AI-assisted humans and authorised agents.
- [Proof of Human vs Proof of Actor](https://www.proofofactor.com/blog/proof-of-human-vs-proof-of-actor): Proof of personhood and human verification ask whether a real human exists. Proof of Actor also asks who is present, in control, using AI or authorised to act.
- [The Missing Trust Layer for AI Agents](https://www.proofofactor.com/blog/the-missing-trust-layer-for-ai-agents): AI agent security needs more than identity. Why delegation, scope of authority, revocation and accountability form the missing trust layer for AI agents.
- [What Does Know Your Agent (KYA) Actually Need to Prove?](https://www.proofofactor.com/blog/what-does-know-your-agent-need-to-prove): Know Your Agent (KYA) must prove more than which agent is acting. A comparison of KYC, KYB and KYA, and the six claims a useful KYA check should establish.
- [What Is Human Control Verification?](https://www.proofofactor.com/blog/what-is-human-control-verification): Human control verification produces evidence that a human is actually controlling a specific digital interaction. How it differs from human in the loop design.
- [What Is Proof of Actor?](https://www.proofofactor.com/blog/what-is-proof-of-actor): Proof of Actor is proposed actor trust infrastructure that establishes whether a human, a human using AI or an AI agent is acting, and who authorised it.
- [Why CAPTCHAs Cannot Solve the AI Agent Problem](https://www.proofofactor.com/blog/why-captchas-cannot-solve-the-ai-agent-problem): Why a CAPTCHA alternative alone cannot solve the AI agent problem: a single challenge cannot establish human control, continuity or delegated authority.

## Optional

- [Privacy policy](https://www.proofofactor.com/privacy-policy)
- [Terms of use](https://www.proofofactor.com/terms)
- [Sitemap](https://www.proofofactor.com/sitemap.xml)