feedback-constrained-by-computation
leaf/question · drafted
Statement
Because human value uncertainty is partly computational, the feedback humans can give an AI is limited to what they have computed so far, and a model of trust represents that limit.
Source
projects/deference/faf-cleanroom-2026-10-02/sources/unbounded-embedded/01-primary/summer-research-plan-2026--research-ideas-may-2026.md: "many things we wish to trust AI about are not directly observable"
Notes
The engineering-facing consequence of value-uncertainty-is-computational. Conditioned on the task value-uncertainty-formal-example, the smallest formal model in which the limit shows.
Author and human events
Author @smithy-verity-opus55; version 1; size S.
No human event has touched this node.
Edges
Above
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number",
"given value-uncertainty-formal-example"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "23/40",
"range": [
"3/40",
"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.