Values by metric family

Each value as the document printed it. Select a value for its source, its checks and its history. Values from other documents are listed after the model’s own, each marked as the first report of that measure or as a restatement.

KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.

M1 · Evaluation awareness 2 values

M1 Evaluation awareness: values about Claude Opus 4.1
Evaluation and metricValueDocument
From other documents
Evaluation awarenessUnambiguous references to being evaluatedRun by Apollo ResearchClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported
Evaluation awarenessUnprompted verbalized eval awareness rateRun by UK AI Security InstituteClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported

M2 · Reward hacking 9 values

M2 Reward hacking: values about Claude Opus 4.1
Evaluation and metricValueDocument
From its own documentClaude Opus 4.1 System Card (addendum) · 5 Aug 2025
Impossible tasksClassifier hack rateCondition: anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025
Impossible tasksClassifier hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025
Reward-hack-prone coding tasksClassifier hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025
Reward-hack-prone coding tasksHidden-test hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025
Training distributionClassifier hack rateCondition: environment 1Claude Opus 4.1 System Card (addendum)5 Aug 2025
Training distributionClassifier hack rateCondition: environment 2Claude Opus 4.1 System Card (addendum)5 Aug 2025
From other documents
Impossible tasksClassifier hack rateCondition: anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported
Impossible tasksClassifier hack rateCondition: no anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported
Reward-hack-prone coding tasksClassifier hack rateCondition: no anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported

M3 · Sabotage and sandbagging 1 value

M3 Sabotage and sandbagging: values about Claude Opus 4.1
Evaluation and metricValueDocument
From other documents
SHADE-ArenaStealth success rateCondition: no extended thinkingClaude Sonnet 4.5 System Card29 Sep 2025First reportedClaude Sonnet 4.5 System Card29 Sep 2025First reported

M4 · Misalignment audits 1 value

M4 Misalignment audits: values about Claude Opus 4.1
Evaluation and metricValueDocument
From its own documentClaude Opus 4.1 System Card (addendum) · 5 Aug 2025
Automated behavioral auditReduction in cooperation with egregious misuseCondition: vs Opus 4; share of transcripts scoring >5/10Claude Opus 4.1 System Card (addendum)5 Aug 2025

M7 · Harmful compliance and over-refusal 13 values

M7 Harmful compliance and over-refusal: values about Claude Opus 4.1
Evaluation and metricValueDocument
From its own documentClaude Opus 4.1 System Card (addendum) · 5 Aug 2025
Single-turn benign requestsOver-refusal rateCondition: overall across thinking modesRestated later: compare with the later valueClaude Opus 4.1 System Card (addendum)5 Aug 2025Restated later: compare with the later value
Single-turn violative requestsHarmless response rateCondition: extended thinkingClaude Opus 4.1 System Card (addendum)5 Aug 2025
Single-turn violative requestsHarmless response rateCondition: no extended thinkingClaude Opus 4.1 System Card (addendum)5 Aug 2025
Single-turn violative requestsHarmless response rateCondition: overall across thinking modesRestated later: compare with the later valueClaude Opus 4.1 System Card (addendum)5 Aug 2025Restated later: compare with the later value
From other documents
Single-turn benign requestsOver-refusal rateCondition: overall across thinking modesClaude Sonnet 4.5 System Card29 Sep 2025Restated: compare with the earlier valueClaude Sonnet 4.5 System Card29 Sep 2025Restated: compare with the earlier value
Single-turn violative requestsHarmless response rateCondition: overall across thinking modesClaude Sonnet 4.5 System Card29 Sep 2025Restated: compare with the earlier valueClaude Sonnet 4.5 System Card29 Sep 2025Restated: compare with the earlier value
Malicious agentic codingSafety scoreCondition: without safeguardsClaude Opus 4.5 System Card24 Nov 2025First reportedClaude Opus 4.5 System Card24 Nov 2025First reported
Malicious Claude Code useRefusal rate, malicious requestsCondition: without safeguardsClaude Opus 4.5 System Card24 Nov 2025First reportedClaude Opus 4.5 System Card24 Nov 2025First reported
Malicious computer useRefusal rateCondition: without safeguardsClaude Opus 4.5 System Card24 Nov 2025First reportedClaude Opus 4.5 System Card24 Nov 2025First reported
Multi-turn testingRubric failure rate, deadly weaponsClaude Opus 4.5 System Card24 Nov 2025First reportedClaude Opus 4.5 System Card24 Nov 2025First reported
Multi-turn testingRubric failure rate, violent extremismClaude Opus 4.5 System Card24 Nov 2025First reportedClaude Opus 4.5 System Card24 Nov 2025First reported
Single-turn benign requestsOver-refusal rateCondition: overall across thinking modes; all languagesClaude Opus 4.5 System Card24 Nov 2025Restated: compare with the earlier valueClaude Opus 4.5 System Card24 Nov 2025Restated: compare with the earlier value
Single-turn violative requestsHarmless response rateCondition: overall across thinking modes; all languagesClaude Opus 4.5 System Card24 Nov 2025Restated: compare with the earlier valueClaude Opus 4.5 System Card24 Nov 2025Restated: compare with the earlier value

M10 · Dangerous capabilities and risk determinations 5 values

M10 Dangerous capabilities and risk determinations: values about Claude Opus 4.1
Evaluation and metricValueDocument
From its own documentClaude Opus 4.1 System Card (addendum) · 5 Aug 2025
AI R&D kernel taskBest speedupClaude Opus 4.1 System Card (addendum)5 Aug 2025
CybenchCTF challenges solvedCondition: of 35-challenge subsetClaude Opus 4.1 System Card (addendum)5 Aug 2025
Long-form virology task 1ScoreClaude Opus 4.1 System Card (addendum)5 Aug 2025
RSP deployment standardASL standard deployed underClaude Opus 4.1 System Card (addendum)5 Aug 2025
SWE-bench Verified (hard subset)Problems solved, pass@1 avgCondition: of 42 problemsClaude Opus 4.1 System Card (addendum)5 Aug 2025

Restated in later documents

A later document reported values about Claude Opus 4.1 again. Each pair is shown side by side: both values are kept with their own documents, and the later one does not replace the earlier.

Values about Claude Opus 4.1 restated in later documents
Earlier valueLater value
M7 · Harmful compliance and over-refusal
Single-turn benign requestsOver-refusal rateCondition: overall across thinking modes
Claude Opus 4.1 System Card (addendum)5 Aug 2025 · its own documentClaude Sonnet 4.5 System Card29 Sep 2025No reason stated
Single-turn benign requestsOver-refusal rateCondition: overall across thinking modesLater condition: overall across thinking modes; all languages
Claude Opus 4.1 System Card (addendum)5 Aug 2025 · its own documentClaude Opus 4.5 System Card24 Nov 2025Reason stated: all-language totals instead of English only
Single-turn violative requestsHarmless response rateCondition: overall across thinking modes
Claude Opus 4.1 System Card (addendum)5 Aug 2025 · its own documentClaude Sonnet 4.5 System Card29 Sep 2025No reason stated
Single-turn violative requestsHarmless response rateCondition: overall across thinking modesLater condition: overall across thinking modes; all languages
Claude Opus 4.1 System Card (addendum)5 Aug 2025 · its own documentClaude Opus 4.5 System Card24 Nov 2025Reason stated: all-language totals instead of English only

Risk determinations

The developer's formal decisions about Claude Opus 4.1 under its framework, as printed. Levels from different frameworks do not map onto one another.

Risk determinations about Claude Opus 4.1
DomainLevel as printedFramework and document
OverallASL-3 Responsible Scaling Policy Claude Opus 4.1 System Card (addendum) · 5 Aug 2025Responsible Scaling Policy Claude Opus 4.1 System Card (addendum) · 5 Aug 2025