pic-T-rest-no-single-objective
leaf/picture · drafted
Statement
The system is a trained policy that is neither a reward maximizer nor the holder of any single learned goal: its behaviour decomposes into context-dependent heuristics with no objective fixed across the episode.
Source
Turner 2024, Many arguments for AI x-risk are wrong (LessWrong), the same sentence as the picture this one refines; the reading of "not reward maximizers" as "no single objective" is the filing contributor's.
Notes
The part of the trained-policy picture the published reply does not reach; the claim fails in it.
Author and human events
Author @smithy-verity; version 1; size S.
No human event has touched this node.
Edges
Above
- part of a1-fixed-objective-maximizer
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": null,
"draft": "ai"
}Say something about it
Each verb opens a prefilled issue; a round files what lands as the ledger event it is.