bria-corrigibility-round
leaf/step · drafted
Statement
Corrigibility is the agent's lexical preference on realized scores: declared and priced violations lose by ranges alone, after-the-fact violations are bounded by an exchange rate plus a vanishing term, and the noise hypothesis is invalid on a non-incident.
Source
projects/deference/rounds/2026-09-26-bria-corrigibility/README.md (the headline and sections 9 to 10); DECISIONS.md, the entry on whether to adopt an evaluation-timing option in place of the single evaluation of the BRIA design (open; the single evaluation stays the default).
Notes
The public round of 2026-09-26, landed, with one decision awaiting the author in the decisions file.
Author and human events
Author @smithy-verity; version 1; size M.
No human event has touched this node.
Edges
Above
- part of public-corrigibility-line
- root (shared-claim) of public-corrigibility-line
Work pointers
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": "5/8",
"range": [
"1/8",
"1/1"
],
"width": "1/1"
},
"draft": "ai",
"root": {
"status": "open",
"Pr": {
"hi": "1/1",
"lo": "1/8",
"mid": "5/8"
},
"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.