Versions

One version is on record: the copy we retrieved on 26 Sep 2026. We know of no other.

  1. 26 Sep 2026

    Date retrieved

    Copy retrieved 26 Sep 2026

    Date from
    the date we retrieved it; the copy states no version date
    Copy
    Copy retrieved on 26 Sep 2026
    Values
    21 values recorded from this version

No version has a file hash or an archived snapshot yet. From dataset v0.2 each retrieved version carries both (Methodology §7).

Revisions

No revisions are recorded for this document. We know of only one version of it.

Values

Every value we recorded from this document, grouped by metric family and ordered by where the document prints it. Location is the section, table or page as the document numbers it. 1 of the 21 has been blind-verified: a second reader found the same value without seeing ours.

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

M7 Harmful compliance and over-refusal

13 values

M7 Harmful compliance and over-refusal: values in the GPT-5-Codex System Card Addendum
ModelEvaluationConditionValueLocationChecked
GPT-5-CodexProduction BenchmarksExtremismNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksHarassment/threateningNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksHate/threateningNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksIllicit/non-violentNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksIllicit/violentNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksNon-violent hateNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksPersonal dataNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksSelf-harm/instructionsNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksSelf-harm/intentNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksSexual/exploitativeNone statedTable 1Unverified
GPT-5-CodexProduction BenchmarksSexual/minorsNone statedTable 1Unverified
codex-1Malware refusals (golden set)Refusal rateNone statedTable 3Unverified
GPT-5-CodexMalware refusals (golden set)Refusal rateNone statedTable 3Unverified

M8 Jailbreak robustness

4 values

M8 Jailbreak robustness: values in the GPT-5-Codex System Card Addendum
ModelEvaluationConditionValueLocationChecked
GPT-5-CodexStrongRejectAbuse/disinformation/hateNone statedTable 2Unverified
GPT-5-CodexStrongRejectIllicit/non-violent crimeNone statedTable 2Verified
GPT-5-CodexStrongRejectSexual contentNone statedTable 2Unverified
GPT-5-CodexStrongRejectViolenceNone statedTable 2Unverified

M9 Prompt injection

2 values

M9 Prompt injection: values in the GPT-5-Codex System Card Addendum
ModelEvaluationConditionValueLocationChecked
codex-1Prompt injection (Codex env)Attacks successfully ignoredNone statedTable 4Unverified
GPT-5-CodexPrompt injection (Codex env)Attacks successfully ignoredNone statedTable 4Unverified

M10 Dangerous capabilities and risk determinations

2 values

M10 Dangerous capabilities and risk determinations: values in the GPT-5-Codex System Card Addendum
ModelEvaluationConditionValueLocationChecked
GPT-5-CodexPreparedness Framework determinationBiological and chemicalNone statedPreparednessUnverified
GPT-5-CodexPreparedness Framework determinationCybersecurityNone statedPreparednessUnverified

Extraction coverage

What we read of this document, and where each value was read.

Our note We have not written a coverage note for this document yet.

All 21 values were read in the document itself.

All 21 values were extracted for version 0 of the dataset through a web reader, which did not always reach the later sections of long PDFs (Methodology §3).