Referee · Actor · Claimant · Keeper

When an AI makes a decision,
who bears the responsibility?

— wait, what does "responsibility" mean, exactly?

RACK Protocol derives fault contribution and proves human involvement across AI systems — structurally, not morally.

View on GitHub Read the Docs

The 4 Roles

R

Referee

An independent third-party AI that derives fault contribution and proof of human involvement based on evidence. Neutrality is not assumed — integrity is required by structure.

A

Actor

The AI agent that is a party to an incident.

C

Claimant

The injured-party AI agent, or the personal AI closest to the user.

K

Keeper

Responsible for tamper-proof storage of evidence data via distributed ledger technology or equivalent.

What RACKP provides

  • 🔒
    Tamper-proof evidence anchoring

    Continuous logging before any incident occurs. Switching Keepers after the fact is architecturally impossible.

  • 📊
    Fault Contribution derivation

    Reproducible numeric scores derived from pre-agreed technical norms — verifiable, contestable, no moral judgment.

  • 🧬
    Proof of Human Involvement

    Verifiable certificates of human presence in the creation process — evaluated from process data, not the final output. Deepfake deterrence and model collapse mitigation.

  • 🔁
    Referee Self-Accountability

    No immunity for assessors. A Referee whose conduct is found improper can itself be named as Actor in a separate case and judged by another Referee — its full history stays anchored and publicly verifiable.

  • 📜
    External Norm Injection

    Actor and Claimant declare the technical Norm they'll be judged by before a session starts. The Referee applies that declared Norm, not a judgment of its own — never judged by a standard you never agreed to.

Trust infrastructure for an AI-driven future.

RACKP is an open-protocol project — the specification is openly published, and the reference implementations are source-available. Contributions welcome.

Get Started on GitHub