rs-press-discounting
leaf/step · drafted
Statement
As a calibrated agent's residual error falls faster than its overseers' false-press rate, its probability of being wrong given a press falls below c/(c+h) and the agent rationally declines to comply with presses.
Source
projects/deference/corrigibility-note-dump-2026-10-01/position-statement.md, section 2.13(b) (the drafting session's sorting: as capability grows the residual error falls and presses become rationally discounted, fully updated deference as arithmetic); the clean-room results erosion_fails_eventually (projects/deference/faf-cleanroom-2026-10-02/packages/corr-three-step-facts) and fails_beyond_level (projects/deference/faf-cleanroom-2026-10-02/packages/corr-legit-modif).
Notes
Threatens: the corrigibility line's plans, the landed deference programme, and the public kernel only where its exchange rate is learned. Formally exhibited in toys. A risk story the corrigibility work itself produced: it shows a probability and a status and enters the comparison through the plans whose tables cite it. No outcome class is tagged (tagging a failure class is a human judgment under the protocol's rule on classes), so its failure mass goes to the default trajectory until a human tags it.
A root of corrigibility-landscape, the sub-frame whose plans it threatens; the top frame keeps only stories at the level of its own question.
Author and human events
Author @smithy-verity; version 1; size M.
No human event has touched this node.
Edges
Above
- part of corrigibility-landscape
- given of crux-oversight-keeps-pace
- root (risk-story) of corrigibility-landscape
- given crux-oversight-keeps-pace
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number",
"given crux-oversight-keeps-pace",
"interval not a bound (negative evidential edge)"
],
"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",
"root": {
"status": "open",
"Pr": {
"hi": "7/8",
"lo": "7/80",
"mid": "37/80"
},
"value": null,
"cost": "85/1",
"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.