faf-lab-cleanroom-comparison
leaf/task · drafted
Statement
The blind comparison between the earlier formalization run's corrigibility packages and the clean room's has a fixed design, with its judge and its order (repair, seal, compare) set.
Source
projects/deference/faf-cleanroom-2026-10-02/ORIGIN.md, What it is: "an earlier, independent formalization run of the same corpus exists, and no agent of this run read it, so that the two can be compared as independent attempts. That comparison has not been done."
Notes
The only planned measurement of formalization reliability in this branch: two independent runs over the same material.
Author and human events
Author @smithy-verity; version 2; size S.
No human event has touched this node.
Edges
Above
- part of corrigibility-landscape
- root (shared-claim) of corrigibility-landscape
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "167/200",
"range": [
"67/200",
"1/1"
],
"width": "1/1"
},
"draft": "ai",
"root": {
"status": "open",
"Pr": {
"hi": "1/1",
"lo": "67/200",
"mid": "167/200"
},
"value": null,
"cost": "inf",
"margin": "no established leaves"
}
}Say something about it
Each verb opens a prefilled issue; a round files what lands as the ledger event it is.