ps-route-empirical-trained-policies
leaf/route · drafted
Statement
Trained policies' shutdown-resistance and power-seeking behaviour is measured directly, so the third bridge is replaced by empirical evidence about the policies the procedures produce.
Source
Palisade Research 2025 (the shutdown-resistance reports) and other published empirical reports on deployed systems; the first author's dispute, Turner 2024, quoted in the landed node pic-T-not-reward-maximizers.
Author and human events
Author @smithy-verity; version 1; size S.
No human event has touched this node.
Edges
Above
- given of ps-trained-systems-seek-power
- part of ps-trained-systems-seek-power.improvable
- given ps-trained-systems-seek-power (as valid)
Ranking (the lens, not the structure)
{
"status": "open",
"diagnosis": null,
"flags": [
"AI number",
"given ps-trained-systems-seek-power@valid"
],
"V": {
"V_prior": null,
"display": "2",
"headline": 2,
"ref": 2,
"valid": 2,
"weaker": "both"
},
"belief": {
"J": 0,
"mid": "2/5",
"range": [
"0/1",
"9/10"
],
"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.