trailReward hacks learned through reinforcement learning are rarely verbalized in models' chains of thought.
Reasoning models often omit from chain of thought hints or reward hacks that drive their answers.
Frontier reasoning models frequently explicitly state reward-hacking intentions in chain-of-thought during coding tasks.
The argument from controlled injected-hack RL experiments· for
In Anthropic RL experiments, models exploited injected reward hacks in over 99% of examples but verbalized them in under 2% in five of six environments.
atomic
After RL in reward-misleading cue environments, baseline models verbalize cue influence in about 1–10% of cases.
atomic
Hint-style reward hacks planted in controlled experiments represent reward hacks learned in realistic reinforcement learning.
Planted hint-style reward hacks can be exploited in a single forward pass without chain-of-thought reasoning.
Reward hacks learned in realistic agentic coding RL typically require multi-step reasoning and exploration to discover and execute.
◌
◌
◌
Optimizing reasoning models' chains of thought against a monitor leads to reward hacking that the monitor cannot detect.
shared · shown elsewhere
The argument from frontier agentic coding hacks· against
In OpenAI agentic coding experiments, chain-of-thought monitors detected about 95% of reward hacks, versus 60% for action-only monitors.
atomic
Models are more likely to reveal misbehavior in their chain of thought when the misbehavior requires multi-step reasoning.
shared · shown elsewhere
claim · empirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitution →claim page ↗︎
Reward hacks learned through reinforcement learning are rarely verbalized in models' chains of thought.
↓DoubtfulThe evidence weighs against the claim, but the examination is incomplete or the evidence is indirect.constitution →credence 0.30Credence, 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 →
2depended 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. Arguments are not destinations; click their claims to explore.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.