The Case for Protocol-Level Proof Systems

Proof can live inside a product or at the protocol layer. Inside a product, it serves that product's users; at the protocol layer, it serves everyone on the same terms.

Product-level proof is often convenient, but its identifiers, access rules and verification logic can disappear when the product changes or the account closes. A counterparty must usually trust the same operator that created the record.

A protocol-level system publishes the record format and settlement rules. Independent clients can derive the same claim identifier, verify the same evidence commitment and read the same outcome without relying on a private database or privileged interface.

Neutrality requires constraints, not slogans. Governance may tune operational parameters, but it should not rewrite individual outcomes or remove hard safety floors at will. Upgrade authority, emergency controls and degraded modes must remain visible and auditable.

The tradeoff is discipline: deterministic canonicalization, cross-platform test vectors, explicit finality rules and portable exports are harder than a closed application workflow. They are also what allow many products to share a proof layer without sharing an operator.