Sourceunknowntext stored

Low-stakes alignment

paulfchristianoDate not recordedOriginal︎ ↗Archived copy

Claims1

ExaminationsNone

Text50 sections

VersionsOne known

CopiesNone recorded

CorrectionsNone recorded

Facts checked Nov 11, 2026

ReachNo copies recorded

StructureIn 1 claim's map0 sources downstream across 1 claim

EvidenceNot yet read

ReviewsNot reviewed

Examinations

None recorded. An examination is commissioned by a claim’s Steward for what its claim rests on; a review of the document for its own sake is asked for as a request bearing on it.

Text103 passages · 1 marked

No examination yetMarked passage

The main substantive claim is that we don’t need to worry about the “distributional shift” between past decisions and future decisions. When the distribution of inputs change, the system may behave poorly for a while, but if we keep retraining on the new data then it will eventually adapt. If individual decisions are low stakes, then the total cost of all of this adaptation is small. I give this argument in more detail in section 7.

Training an AI system on an accurate oversight signal makes it act as its overseers intend throughout deployment, including in rare high-stakes situations.Doubtfulmentions

Not yet read by this claim’s Steward.

Lineage across 1 claim

Draws on · none recorded

This source

Low-stakes alignment

in 1 claim's map

Drawn on by · none recorded

Record

Kind: unknown
Authors: paulfchristiano
Publisher: Not recorded
Published: Not recorded
DOI: None
Link: https://www.alignmentforum.org/posts/TPan9sQFuPP6jgEJo/low-stakes-alignment
Archived copy: https://web.archive.org/web/20260807111138/https://www.alignmentforum.org/posts/TPan9sQFuPP6jgEJo/low-stakes-alignment
Stored text: 24,727 characters, retrieved Nov 11, 2026 · 50 sections, 103 passages
Claims extracted: Not recorded
Versions: One known
Copies: None recorded
Corrections: None recorded
Facts checked: Nov 11, 2026

History 3

Nov 11, 2026DividedSegmenter · Divided into 152 parts by its own structure
Nov 11, 2026FactsFacts fetcher · Facts looked up
Nov 11, 2026Entered the graphEntered the graph; copy stored