The argument from delegated judging
The argument
The human's part in amplified oversight can be imitated by a judge model trained on human judgments, so it is not the case that a human judgment is needed for each signal. The remaining cost per action is a bounded multiple of inference compute, which falls as hardware and protocols improve, and which a developer that gave safety priority could pay on every action. Per-action amplified oversight is therefore affordable in training and in deployment.
Does it work?
- Does it follow?
- slight even if the premises holdEven if the premises are true, they make the conclusion only a little more likely.constitution →
- Do the premises hold?
- 1 not separately assessed
Granting its premises, the argument shows that the per-action cost is bounded and could be paid by a developer willing to serve several times fewer actions for the same compute. That removes the strong sense of "too costly" and leaves the weaker one standing. It also partly changes the subject: a judge model that imitates human judgment is what the literature treats as the cheaper substitute for amplified oversight, weaker than the system it oversees, so paying for it on every action is not the same as obtaining the amplified signal on every action. The argument depends on the denial of the premise that each signal needs a human judgment, which is unassessed and on which the defining texts are themselves equivocal.
Premises
- UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitution →Amplified oversight methods such as debate require a human judgment for each oversight signal they produce.
Where it is made
Passages where sources make, criticize or mention this argument. They record where it appears, not how strong it is, and none has precedence over another.
No source is recorded as making it.