experiment: confirm live explicit sufficiency closure

This commit is contained in:
2026-08-15 12:43:56 +01:00
parent abc01b181f
commit a599922d9f
2 changed files with 79 additions and 0 deletions
+21
View File
@@ -4372,3 +4372,24 @@ Observed:
What this proves: The deterministic stripping gate from 60B.80 prevents premature model-led terminal closure of parent decisions when no explicit sufficiency confirmation is present, and allows State B to fire naturally through the existing questioning path.
**Doc:** docs/experiment-60b81.md
### Experiment 60B.82 — Live Explicit Confirmation Closure Confirmed (2026-08-15)
**Type:** Paired live observation — explicit-confirmation positive branch (1 update call, production route)
**Input:** Answer with explicit sufficiency confirmation: *"The enterprise customer has now confirmed in writing that they will not sign if we launch this year, so the £700,000 of expected annual revenue from them will not be received. There are no other material uncertainties between launching this year and waiting twelve months."*
**Classification: A — LIVE EXPLICIT-CONFIRMATION CLOSURE CONFIRMED**
Observed:
- Customer resolved ("Will not sign", negative meaning preserved)
- Decision terminal (status=resolved, value=null)
- Both nodes resolved: ["n_enterprise_customer_signing", "n_product_launch_decision"]
- finalActiveUnknownNodeId = null
- finalSelectedQuestion = null
- No new nodes or edges introduced
- Financial revision correctly applied (£1.2M → £500k reflects confirmed £700k loss)
What this proves: Explicit sufficiency confirmation flows cleanly through the 60B.80 gate to terminal closure of the parent decision, active target clears, and no question is asked. The paired lifecycle (no confirmation→State B / explicit confirmation→terminal close) is live-baselined in production.
**Doc:** docs/experiment-60b82.md
+58
View File
@@ -0,0 +1,58 @@
# Experiment 60B.82 — Live Explicit Confirmation Closure Confirmed
**Date:** 2026-08-15
**Branch:** `feature/decision-closure-ownership-v0.47`
**Preceded by:** Experiment 60B.81 (live no-confirmation State B path confirmed)
**Type:** Paired live observation — explicit-confirmation positive branch
## Objective
After 60B.81 proved the negative branch (no confirmation → decision stays open + sufficiency question), answer: does explicit sufficiency confirmation still close the decision cleanly?
```text
final represented factor resolves
+
explicit sufficiency confirmation present
=>
parent decision may close
=>
active target clears
=>
no further question
```
## Method
- **Fixture:** `tests/fixtures/pre-anchored-product-launch-customer-signing.json`
- **Input (ANSWER_2):** `"The enterprise customer has now confirmed in writing that they will not sign if we launch this year, so the £700,000 of expected annual revenue from them will not be received. There are no other material uncertainties between launching this year and waiting twelve months."`
- **Crucially present:** `"There are no other material uncertainties"` — explicit sufficiency confirmation
- **Mode:** updateOnly (1 update call via production HTTP route)
## Result
**Classification: A — LIVE EXPLICIT-CONFIRMATION CLOSURE CONFIRMED**
### Observed
| Field | Value |
|-------|-------|
| Customer status | resolved |
| Decision status | resolved (terminal) |
| resolvedUnknownNodeIds | ["n_enterprise_customer_signing", "n_product_launch_decision"] |
| finalActiveUnknownNodeId | null |
| finalSelectedQuestion | null |
| addedNodes | [] |
| addedEdges | [] |
| Financial revision | £1.2M/year → £500k/year (correctly reflects £700k enterprise loss) |
### What This Proves
1. Explicit sufficiency confirmation ("no other material uncertainties") flows cleanly through the 60B.80 gate to terminal closure of the parent decision.
2. Both the paired negative branch (60B.81: no confirmation → State B) and positive branch (60B.82: explicit confirmation → terminal close) are now live-baselined.
3. The confirmation-gated lifecycle is symmetric and complete in production.
### What Remains Unproven by This Run
- Edge-case confirmation phrasings (implicit, partial, or negated confirmation language).
- Multi-node simultaneous resolution with confirmation.
- Confirmation under contradiction reasoning constraints.