power-bounded-by-wisdom
leaf/step · drafted
Statement
A rule on the process caps an agent's capability by how well it understands and is motivated by the general welfare, read through an external measure of wisdom.
Source
The author, projects/deference/corrigibility-note-dump-2026-10-01/position-statement-v2.md, section 1.3: "In order for the future to go well, or at least for it to reliably go well, we'll need to increasingly respect rules that upper-bound an agent's power by its wisdom." and "somehow agent capabilities should be capped as a function of how well the agents actually understand and are motivated by the common good."; checklist condition J7 in projects/deference/corrigibility-note-dump-2026-10-01/position-statement-v2.md; projects/deference/corrigibility-note-dump-2026-10-01/corrigibility-bli-thread-and-early-questions-audit.md, row 2.14 (the wisdom meter is gameable); the clean-room result in projects/deference/faf-cleanroom-2026-10-02/packages/corr-power-channel, statement T7(c): erosion of the humans' grip is power-seeking in the one-shot value-certain form.
Notes
Condition J7, read as a rule respected in practice. The meter problem is open; the BLI thread's selection argument says a learned meter selects for undetectable manipulation.
Author and human events
Author @smithy-verity; version 1; size M.
No human event has touched this node.
Edges
Above
- part of dynamic-conditions-for-going-well
- root (shared-claim) of dynamic-conditions-for-going-well
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": "3/20",
"range": [
"0/1",
"13/20"
],
"width": "1/1"
},
"draft": "ai",
"root": {
"status": "open",
"Pr": {
"hi": "13/20",
"lo": "0/1",
"mid": "3/20"
},
"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.