# Provenance: Whose These Are, And What Class Of Evidence They Carry

*Partner artefacts folder, document 00. **Nothing in this folder is this estate's work**, and nothing in it has been verified by this estate.*

---

## What is here

Four artefacts produced by the **RiskMandate business partner** on or around **31 August 2026**, after an in-person meeting with the project lead on 31 August, and **in parallel with** the project lead's memo series (briefs v0.33.71–83). Neither side saw the other's work while producing it.

| Artefact | Audience | In this folder |
|---|---|---|
| **riskmandate.ai landing page** | Public | **Yes** — byte-exact, 37,880 bytes |
| **Design Partner Deck**, 11 slides | Prospects | **Yes** — the PDF, plus a machine text extraction |
| **RAMM-I Insurability Framework v0.1**, 9 pages | Internal | **No — held.** See below |
| **RAMM-I Deck**, 17 slides | Internal | **No — held.** See below |

## Two are held, and why

The two RAMM-I documents are **internal strategy**. They name an existing customer under an active statement of work, name specific acquisition and partnership targets, state a negotiating position on a contract clause, and characterise named competitors in terms written for a private audience.

**This site is public and serves every file in the repository**, linked or not — so committing those PDFs would publish them, and git history would keep them after any deletion.

> **Publishing another company's internal strategy, naming their customer, is the project lead's decision and their partner's, not this estate's.** They are staged and can be added in one commit on a word from the project lead.

Their **doctrinally relevant substance** is carried in [document 01](concordance.html) instead, with customer names, target lists and competitive characterisations deliberately omitted.

## The evidence class of everything in this folder

This matters more than anything else on this page, and it is the rule the estate already holds:

> **A connector labels the evidence class of everything it imports** ([GM-D57](../packs/grant-and-mandate/change-control.html)), or *integrate with what exists* quietly becomes *launder what exists*.

These artefacts are a connector. So:

| | Class | Meaning |
|---|---|---|
| That the partner **asserts** these things | `observed` | The documents are here and say what they say |
| The market claims **inside** them — filings, carrier names, approval rates, form numbers, loss figures | **`documented`** | **A third party's assertion. This estate has verified none of them** |
| That any assertion is **true** | — | **Unestablished** |

**Marked, not merged.** [Document 01's two-channel rule](../insurance/the-rating.html) says a declared fact is discounted and rendered as declared. A market claim arriving in a partner's deck is a declared fact about the world, and treating it as measured because it is specific and well-formatted is precisely the error the rule exists to prevent.

**Several claims here are checkable** — ISO form numbers, regulatory filing approvals, carriers' published AI products, published standards. Checking them would move those rows from `documented` to `read` or `observed`. **Nobody has done that**, and until somebody does, this folder is a record of what a partner said, not of what is true.

## What this does to the folder next door

[The insurance folder](../insurance/index.html) has said since memo 9 that the pivot has produced no external evidence, and `insurance/llms.txt` still leads with it:

> *NO EXTERNAL EVIDENCE HAS BEEN GATHERED … the survey would be the first thing here that could be wrong in a way the world would correct.*

**That statement is still true and is now more interesting.** These artefacts are the *shape* of the survey memo 9 asked for, arrived at independently — but a survey whose findings nobody in this estate has checked is a set of `documented` claims, not gathered evidence.

> **The survey has not been run. Somebody else's survey has been received.** The two are different, and collapsing them would be the first dishonesty in this folder.

## What this does not prove

- **That any market claim here is accurate.** Not one has been checked against a primary source.
- **That the partner's framework and this estate's doctrine agree.** [Document 01](concordance.html) maps where they converge and where they collide; the collision on the scale is real and unresolved.
- **That the convergences are independent confirmation.** Both bodies of work were produced by the same two people after meeting, and by language models, in the same week. Agreement between them is weaker evidence than it looks.
- **That these are final.** The landing page and deck are dated artefacts of one week and the partner may already have moved.

---

*CC BY 4.0 applies to this estate's commentary only. **The artefacts themselves are the RiskMandate partner's and are reproduced here as received.***