corrigibility-faithful-plus-legit-on-self
leaf/step · drafted
Statement
An AI is corrigible to a human process when it is faithful and legitimate on the family of governance questions about modifying it, and governance decisions made by a process sealed from the AI's contemporaneous outputs are the design corollary.
Source
projects/deference/note-dump-2026-08-11/notes/legitimacy-theory-v1.md, sections 6.1 and 6.2 (corrigibility, derived); projects/deference/faf-cleanroom-2026-10-02/packages/legit-li-register/legit-li-register-report.md, the definition CorrigibleOn and the corollary governance_gDiag_defect_not_vanishing.
Notes
The clean room's definition of record for corrigibility in this model, with the diagonal corollary proved; the corrigibility branch holds the reading of it as a corrigibility claim. The open closure-under-selection statement is its action. Shared with corrigibility-risk; the merge keeps one.
Author and human events
Author @smithy-verity; version 1; size M.
No human event has touched this node.
Edges
Above
- held by legitimacy
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number",
"held by legitimacy"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "1/2",
"range": [
"0/1",
"1/1"
],
"width": "1/1"
},
"draft": "ai"
}Say something about it
Each verb opens a prefilled issue; a round files what lands as the ledger event it is.