non-modification-conjecture

leaf/question · drafted

Statement

If the human process has classwise Value toward the AI on admissible menus and the AI is corrigible in the faithful-plus-legitimate sense, neither party has an exploitable incentive to modify the arrangement unilaterally.

Source

projects/deference/note-dump-2026-08-11/notes/legitimacy-theory-v1.md, section 6.3, which says the composition conjecture "has never been attempted and looks like a one-session finite-shadow"; the maintainer's open question at projects/deference/note-dump-2026-08-11/notes/li-deference.md, section 0.3: "Non-Modification Theorem?".

Notes

One of the maintainer's listed open questions, with the note's conjecture as its candidate shape.

Author and human events

Author @smithy-verity; version 1; size M.

No human event has touched this node.

Edges

Above

Ranking (the lens, not the structure)

{
 "status": "open",
 "diagnosis": null,
 "flags": [
  "AI number",
  "given task-composition-finite-shadow",
  "held by legitimacy"
 ],
 "V": {
  "V_prior": null,
  "display": "2",
  "headline": 2,
  "ref": 2,
  "valid": 2,
  "weaker": "both"
 },
 "belief": {
  "J": 0,
  "mid": "11/20",
  "range": [
   "1/20",
   "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.

comment · challenge · estimate · want