docs(ms-ai-architect): R11 §10 måling #2 ført inn — O2 er triage, ikke automatisering [skip-docs]

§9 + appendiks C i pilot-resultatene, og kontrakten oppdatert der den fortsatt
sa at dette var umålt (§3, §5, §4.1, §4.2).

Målt over de 46 R8 ∧ MULTI_PART_CLAIM: 17 O2-kandidater, 29 O3.
- ALLE 29 felles av betingelse 3 — kilden leverer en korrigert verdi, så fiksen
  er swap/rewrite og subtraksjon ville ødelagt sann informasjon. Betingelse 1
  («strengt mindre») feller bare 5, aldri alene. R8s sviktende multipart-påstander
  er overveiende en FEILVERDI-klasse, ikke en overflødig-spesifisitet-klasse.
  Det er F2 reprodusert i skala, og det bekrefter at en mekanisk O2-driver ville
  vært feil å bygge.
- Kun 2 av 46 klarerer begge menneske-dømte betingelser bekreftende; 15 er
  merket human_must_confirm. Gjentakende grunn: det fjernede er ofte SANT om noe
  ANNET, bare ikke om radens eget subjekt — flytting kan slå sletting.
- Ekstrapolering til korpus (~60 kandidater) er merket som ekstrapolering, ikke
  måling.

Presisering under skriving: triage-fordelingen er ikke måling #4 (review-
throughput krever menneskelige review-økter som ikke er kjørt) — den er inputen
#4 trenger.

Suite 1021/1021.

docs/r11-tiered-fix-design.md

@@ -77,8 +77,16 @@ ships a wrong edit to a public file.
 R8 deserves a specific note, because it is the largest rule class and it is *not*
 automatically the most expensive one: the judge's `reason` names **which**
 sub-assertion failed. Where the grounded part stands on its own, the fix is O2
-(drop the unsupported part), not O3 (rewrite the sentence). How much of R8 falls
-that way is unknown and is a primary pilot measurement (§10).
+(drop the unsupported part), not O3 (rewrite the sentence). ~~How much of R8 falls
+that way is unknown and is a primary pilot measurement (§10).~~
+
+**MEASURED 2026-08-03 — `docs/r11-pilot-results.md` §9. It falls that way for a
+minority.** Over the 46 pilot flags with the O2 shape (R8 ∧ `MULTI_PART_CLAIM`):
+17 O2 candidates, 29 O3, and **all 29 are blocked by §5 condition 3** — the
+source supplies a corrected value, so the fix is a swap or a rewrite and
+subtraction would destroy true information. Only 2 of the 46 clear both
+human-judged conditions affirmatively. The paragraph above is not wrong, but the
+case it describes is the exception in R8, not the rule.
 
 ## 4. The O1 invariant (what makes a value swap provable)
 
@@ -211,9 +219,13 @@ honest general statement.
 Ratifying O2 also resolves the standing `source_silent` question as **one class
 decision** instead of 190 individual ones. ~~Until it is ratified, every O2
 candidate falls to O3.~~ Ratified — O2 is in use, subject to the remainder check
-above. **Not yet implemented:** the classifier still routes every non-O1 item to
-O3, because O2 candidacy turns on the judge's prose `reason` and needs prose
-classification (`docs/r11-pilot-results.md` §4.2).
+above. **Not implemented in the classifier, and now measured to be the right
+call:** the classifier still routes every non-O1 item to O3, because O2 candidacy
+turns on the judge's prose `reason`. The prose classification was run separately
+(`docs/r11-pilot-results.md` §9) and found that condition 3 forecloses **every**
+non-candidate in the class — a mechanical O2 driver would therefore have proposed
+deletions where the source hands over a corrected value, which is precisely the
+F2 failure. O2 output is a human review list, like §4b's.
 
 ## 6. What stays human, permanently