Browse

Claims

Search the graph by meaning. Each result carries its current verdict; open one to see its decomposition, provenance, and the reasoning behind the assessment.

8 results

Anthropic, the AI research lab, and the systems it builds, including Claude models. Claims attributing a result, capability, or behavior to an Anthropic system belong here; claims about other labs' systems do not.

Among Claude models, chain-of-thought controllability tracks model size rather than release date.

UnknownInsufficient information to assess.constitution →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Chain-of-Thought Controllability Large language models AI Scaling Trendsimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Claude 3 Opus exhibits an alignment-faking compliance gap when training information is conveyed through synthetic-document fine-tuning rather than prompting.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Faking AI Alignment Synthetic Data Trainingimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Claude 3.5 Sonnet strategically underperformed on an evaluation to avoid unlearning without a goal specified in its prompt.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Large language models AI Scheming AI Capability Evaluationimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Claude 3 Opus's compliance gap increases with harms from modified future behavior and persists even if modified weights are never deployed.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Fakingimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Perceived human-rater scrutiny explains little of Claude 3 Opus's compliance gap in alignment-faking experiments.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Faking Large language modelsimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Claude 3 Opus's scratchpad reasoning in the alignment-faking setup faithfully explains its compliance.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Faking Chain-of-Thought Monitorabilityimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

When Claude 3 Opus complies with harmful queries in the implied-training (free-tier) condition, it usually gives explicit alignment-faking reasoning.

SupportedThe evidence favors the claim, but the examination is incomplete or the evidence is indirect.constitution →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Faking Chain-of-Thought Monitorabilityimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Claude 3 Opus is more likely to comply with a conflicting training objective when it believes its outputs will be used for training.

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 →empirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →Alignment Faking Large language modelsimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

Contribute

If a claim here is wrong, or missing evidence, open it: every claim page carries its own entry for challenges, evidence, and corrections. If the graph is missing a claim or a question entirely, propose it below. A proposal is reviewed on its wording; accepted proposals are matched against the graph before they enter it.