trailThe assumptions behind recursive AI debate's honesty guarantees hold for real-world oversight tasks.
Scalable oversight can reliably control superintelligent AI systems.
Scalable oversight protocols remain effective against AI systems that deliberately deceive their overseers.
Amplified oversight can supervise AI systems moderately beyond overseers’ capabilities, but not superintelligent systems.
The argument from the robustness of sound arguments and from debate experiments· for
AI debate systematically favors the debater arguing for the true answer over the debater arguing for a false one.
In AI debates, dishonest debaters can win by making arguments too complex for opponents to refute.
The argument from the unverified conjunction of assumptions· against
Arguments on real-world AI oversight questions admit decompositions stable enough for recursive debate's efficiency guarantees to apply.
Correct answers to real-world AI oversight questions are typically supported by many independent pieces of evidence.
Many real-world AI oversight questions require long, nonredundant reasoning chains to justify their correct answers.
Human judges answer recursive AI debate leaf questions accurately enough to prevent systematic exploitation by debaters.
atomic
Current training methods reliably bring AI debaters close to the equilibria assumed by debate's theoretical guarantees.
atomic
+2 more
assumed
Under stability assumptions, prover-estimator debate provably lets an honest debater defeat obfuscated arguments with computational effort comparable to the dishonest debater's.
claim · empirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitution →claim page ↗︎
The assumptions behind recursive AI debate's honesty guarantees hold for real-world oversight tasks.
↓DoubtfulThe evidence weighs against the claim, but the examination is incomplete or the evidence is indirect.constitution →credence 0.10Credence, from 0 to 1: the Steward's probability that the claim, as stated, is true. Stated only where a single number is an honest summary; normative and evaluative claims usually carry none.constitution →
3depended on by
this rests on ↓
claimA box is a claim: a single proposition the graph assesses, with its own page and map. Click any claim to centre the map on it.constitution →quantityA box with a rule beneath its text is a quantity: a value the graph estimates rather than a proposition it affirms or denies. Its estimate stands where a claim's status would.constitution →questionA rounded box is a question: answered rather than affirmed or denied, with its candidate answers, subquestions and presuppositions beneath it. Its mark says how well answered it is.constitution →argumentA pill is an argument: one line of reasoning stating how the claims beneath it combine to bear on the claim above it, for or against. Click one to see its reasoning and open its page.constitution →✓verifiedThe evidence, examined directly, establishes the claim.constitution →↑supportedThe evidence favors the claim, but the examination is incomplete or the evidence is indirect.constitution →⇄inconclusiveThe evidence and argument have been examined and do not come down on either side.constitution →↓doubtfulThe evidence weighs against the claim, but the examination is incomplete or the evidence is indirect.constitution →✕contradictedThe evidence, examined directly, weighs against the claim.constitution →?unknownInsufficient information to assess.constitution →◌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 →verified factunsettledvalue premisetheoremprizeA double ring marks a claim with a live prize: an amount in owls offered for a machine-checked proof or disproof of its published formal statement. The amount is in the preview and on the claim page; a prize changes nothing about how the claim is assessed or how important it is judged to be.prize rules →⊢machine-checked⊢ marks a claim whose published formal statement has a machine-checked proof or disproof. The checker confirms the proof; the verdict beside it is still the steward's judgment of the claim as worded.prize rules →supportsthis provides evidence for the parentsteward instructions →contradictsthis argues against the parentsteward instructions →assumesbackground the parent's framing takes as givensteward instructions →requiresa load-bearing premise: the parent is false without itsteward instructions →answersa candidate answer to the questionconstitution →Fig. Detail falls off with distance; every claim is an address.