values-ahead-of-capability
leaf/step · drafted
Statement
For each action that becomes available at a step of the trajectory, the agent's expected harm from that action exceeds its value of information before the action is available.
Source
The author, projects/deference/corrigibility-note-dump-2026-10-01/position-statement-v2.md, section 1.3: "Training needs to focus on value learning 'ahead of schedule' ie before it is needed, so that the value of information of unkind human experimentation is outweighed by the disutility."; the drafting session's formalization as checklist condition J3 in projects/deference/corrigibility-note-dump-2026-10-01/position-statement-v2.md; run 3's most-likely-wrong item 5 in projects/deference/faf-cleanroom-2026-10-02/sources/corrigibility/workflow-2026-09-14b/plan/synthesis.md: the condition is conjectured, measured by nothing built, violated in the one deployed anchor.
Notes
Condition J3, read of the current default training trajectory. Nothing measures value coverage; the action is to define one.
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": "1/5",
"range": [
"0/1",
"7/10"
],
"width": "1/1"
},
"draft": "ai",
"root": {
"status": "open",
"Pr": {
"hi": "7/10",
"lo": "0/1",
"mid": "1/5"
},
"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.