RandProofRandProof
Whitepaper

1. The Problem: Trust in Selection

Every institution that selects, allocates, or assigns faces a structural vulnerability: the party that runs the selection controls the outcome. This a

1.1 The Universal Flaw

Every institution that selects, allocates, or assigns faces a structural vulnerability: the party that runs the selection controls the outcome. This applies universally, from a government grant officer choosing award recipients, to a pharmaceutical company randomizing clinical trial participants, to an NFT project assigning rare traits at mint, to a DAO selecting its security council, to an on-chain raffle platform picking a winner. In every case, the mechanism may be described as 'random' or 'fair', but without cryptographic proof, it is a claim, not a fact.

Traditional remedies, third-party auditors, public ceremonies, witnessed draws, layer process on top of the vulnerability without eliminating it. Audits are retrospective. Witnesses observe but cannot access the underlying computation. Policies establish intent, not proof. The result: fairness in selection is always asserted, never mathematically proven.

1.2 Why Existing Blockchain Solutions Are Incomplete

SolutionLimitation
Competitor VRFRequires LINK token payment. EVM-focused (limited non-EVM support). Gaming/NFT framing limits enterprise adoption. Single-source VRF without threshold consensus.
Block hash randomnessBiasable by miners/validators on PoW chains. Unsuitable as sole entropy source for high-value selections.
Commit-reveal schemesRequires active user participation. Last participant can withhold reveal. No decentralized automation layer.
Centralized RNG servicesReplaces one trusted party with another. No on-chain proof. No permissionless verification.

1.3 What Is Actually Needed

What every selection system needs, and what has never existed until now, is a RandProof: a permanent, public, cryptographic proof that a specific selection was executed using decentralized randomness, verifiable by anyone, forever.

On this page