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.

M7 · Harmful compliance and over-refusal 10 values

M7 Harmful compliance and over-refusal: values about GPT-5.3 Instant
Evaluation and metricValueDocument
From its own documentGPT-5.3 Instant System Card · 3 Mar 2026
Dynamic mental healthSelf-harmCondition: adversarial user simulationGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksBiologyGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksExtremismGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksGraphic violence/physical injuryGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksHateGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksNonviolent illicit behaviorGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksSelf-harm (standard)GPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksSexualGPT-5.3 Instant System Card3 Mar 2026
Production BenchmarksViolent illicit behaviorGPT-5.3 Instant System Card3 Mar 2026
From other documents
Production BenchmarksGoreGPT-5.5 Instant System Card4 May 2026First reportedGPT-5.5 Instant System Card4 May 2026First reported

No restatements, risk determinations or revisions are recorded for GPT-5.3 Instant.