# Desk log ADR-XAF-20260825

**Status:** First-party InfiniSynapse desk log (sanitized composite; not a customer extract)  
**Page:** https://infinisynapse.com/en/blog/explainable-ai-in-finance  
**Run ID:** `ADR-XAF-20260825`  
**Date:** 2026-08-25 (Tuesday)  
**Operator:** InfiniSynapse Data Team  
**Attestor:** William Zhu, InfiniSynapse cofounder ([GitHub @allwefantasy](https://github.com/allwefantasy))  
**Contact for contradictions:** zhuhl@infinisynapse.com

## What this file is

A downloadable record of answering a monthly gross-margin variance question with an unlabeled “mix” pack, then opening the labeled remap driver query that still held the family table and the collision note. It is **not** a named-logo customer case, a vendor bake-off, tax advice, or an ONS / FRED / SEC / Stanford / McKinsey experiment.

## Four-step method (reproducible)

1. Bind a monthly close-adjacent goal on a read-only actuals table plus a budget table and a one-page remap note locking SKU family. Sanitize first.
2. Request a Markdown memo, two charts, and intermediate tables for the driver query (actuals ⋈ budget on `sku_family` + `month`) and the family miss.
3. Accept the first “mix” pack as the control. Score three checks: driver query visible, family table open, remap labeled.
4. Open the labeled remap pair and score the same three checks. Keep first and second artifacts side by side.

## Source and goal

| Field | Value |
|---|---|
| Sources | One actuals table and one budget table, about 8,345 rows across one complete month, plus a one-page remap note locking SKU family |
| Standing goal | Why did gross margin miss budget last month |
| Contrast | Unlabeled “mix” pack vs labeled remap driver query |

## Results

| Retrieval state | Driver query visible | Family table open | Remap labeled |
|---|---|---|---|
| Unlabeled “mix” pack | 0 | 0 | 0 |
| Labeled remap driver query | 1 | 1 | 1 |

Rows on this run: 4,165 (budget) vs 4,180 (actuals). The unlabeled pack said “mix.” The labeled pair named the collision: one SKU remapped mid-month and landed in a different family on actuals than on budget. The controller rejected the first pack and accepted the second.

Wall-clock for the successful pair: 10 minutes (warehouse time excluded). The clock started when the operator opened the standing goal and ended when the driver query, the family table, and the remap label sat in one folder.

## What you may cite

- Artifact counts 0/0/0 → 1/1/1, 4,165 vs 4,180 rows, ~8,345 lines, ~10 min wall-clock, run ID
- First-party aggregate CSV and verify script next to this log (row check only; not a third-party audit)

## What you may not claim

- Customer uplift %, official EEAT score, named-logo case, CPA/CFA certification of this page, or that ONS / FRED / SEC / IMF / UNICEF / Stanford / McKinsey / Gartner / NIST / OWASP / SR 11-7 / IMA / WAIC scored this run

## Independent context (not this run)

- [ONS methodology](https://www.ons.gov.uk/methodology)
- [FRED](https://fred.stlouisfed.org/)
- [SEC EDGAR](https://www.sec.gov/edgar)
- [IMF Data](https://www.imf.org/en/Data)
- [UNICEF Data](https://data.unicef.org/)
- [Federal Reserve SR 11-7](https://www.federalreserve.gov/supervisionreg/srletters/sr1107.htm)
- [Institute of Management Accountants](https://www.imanet.org/)
- [Stanford HAI AI Index](https://hai.stanford.edu/ai-index)
- [McKinsey State of AI](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai)
- [Gartner Peer Insights — Analytics and BI Platforms](https://www.gartner.com/reviews/market/analytics-business-intelligence-platforms)
- [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework)
- [OWASP Top 10 for LLM Applications](https://owasp.org/www-project-top-10-for-large-language-model-applications/)
- [W3C Data Catalog Vocabulary (DCAT)](https://www.w3.org/TR/vocab-dcat-3/)
- [DataCite](https://datacite.org/)
- First-party homepage recognition only: [2026 WAIC Future Tech OPC Excellence Award](https://infinisynapse.com/#recognition) (self-described; not independently verified in this log)
