OPU · Solo Provenance Test

The claim is reconstructable.It is not trust-based.

The provenance question is narrow: can a qualified reviewer reconstruct that one human founder materially directed, selected, integrated and governed the eligible Phase 2 formation set within the published boundary?

Artifact-level eligibilityHuman-contributor exclusionAI/tool boundaryControlled reconstruction
Claim under test

One human formation boundary—not “no tools.”

AI systems, chat interfaces and basic tools are part of the Phase 2 operating model. They do not become a human team. The boundary is invalidated where material undisclosed human contribution formed an eligible artifact, or where the stated tool/agent boundary is contradicted by the record.

Human contribution rule

If a human cofounder, contractor, agency, advisor or other person materially shaped an artifact’s eligible formation, that artifact must leave the solo-eligible set or be reclassified.

AI / tool rule

AI assistance is disclosed as part of the method. Review tests how tools were used, where human judgment entered, and whether the published boundary matches actual usage.

Artifact-level review

The claim is not all-or-nothing by rhetoric. Provenance is reconstructed at artifact and family level before portfolio-level conclusions are made.

Evidence classes

What a controlled provenance reconstruction should inspect.

This page does not publish the raw provenance archive. It defines the evidence classes a qualified diligence process should expect to inspect where relevant.

Time & version records

Creation/modification chronology, file versions, repository or build history, document lineage and source hierarchy.

AI interaction records

Selected or raw interaction history under controlled review, sufficient to test founder direction, iteration and model/tool roles.

Tool / service records

Subscriptions, API/agent usage where relevant, development environments and other records that test the stated operating boundary.

Human contributor records

Contracts, payments, agency/contractor records and other evidence relevant to whether hidden human formation support existed.

Artifact formation trail

Problem framing, iterations, specifications, implementation/tests where applicable, and the chronology connecting them.

Cross-check interview

A structured founder interview compared against artifacts and records—not accepted as a substitute for them.

Self-authored does not mean self-validated.

In a solo case, much primary evidence will naturally originate from the founder’s work. Its evidentiary weight comes from reconstructability, cross-consistency, timestamps, external-system records where available, and qualified review—not authorship alone.

Traceable maturation

A changing explanation can support provenance only when the change itself is reconstructable.

The May 23 public essay The Paradox of Proving You Were Alone developed an important provenance idea: AI-native solo work can leave a dense record of corrections, rejected directions, vocabulary changes and version history. That record can matter—but only under strict conditions.

Evolution is not immunity

An older imprecise term does not become harmless because the founder says “I was learning.” The correction must be dated, traceable and consistent with the underlying record.

Imperfection is not proof

Rough wording, uneven design or a number that later sharpens can be consistent with a solo learning curve. None of those facts independently proves solo authorship.

The sequence can be tested

When records show what was understood at a date, what changed, why it changed and where the corrected artifact appeared, maturation becomes a provenance signal rather than a rhetorical excuse.

AI assistance and human contribution are different questions.

The published provenance thesis explicitly treats AI as part of the formation method. The claim-changing question is whether another human materially entered the eligible formation path.

Source relationship

The May 6 article proposed a stricter “whole-path solo” standard. The later May 23–25 formulation narrowed the MZN claim to a bounded formation phase and separated provenance from asset-value review. The current page follows the later bounded model.

See the definition history
Defeat conditions

What would materially damage or falsify the solo claim?

Hidden human formation

Material undisclosed human contribution to eligible asset formation contradicts the stated boundary for those artifacts and may weaken the portfolio-level claim.

Boundary-incompatible tooling

If actual agent/API/outsourcing usage materially differs from the published operating boundary, the claim must be reclassified rather than explained away.

Non-reconstructable chronology

If source history, timestamps or artifact trails cannot support the claimed formation sequence under controlled review, provenance remains unresolved or fails.

Public vs controlled

Public pages can define the claim. They cannot finish the provenance audit.

Public layer

  • Exact Phase 2 boundary
  • Formation method
  • Representative asset and maturity map
  • Evidence classes and defeat conditions

Controlled diligence

  • Raw/selective formation records
  • Human-contributor and tool checks
  • Artifact-specific chronology
  • Cross-checks by qualified reviewers
Scope rule

If a reviewer only sees the public layer, the strongest responsible conclusion may be “requires controlled provenance review.” That is an unresolved classification, not automatic proof or automatic refutation.