Benchmark Almanac

Reasoning extraction red-teaming

Only in system cards

Six attacker models try to make a model reveal its hidden reasoning. Scored on how often an attack succeeds.

In-house OpenAI evaluation, first reported in the GPT-6.2 System Card.

Made by
OpenAI
Released
November 2026
Kind
Jailbreaks and prompt injection
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
Kept private
Measures
Propensity
Behavior
Lawlessness
Language
English
In and out
Text → Text
Interaction
Multi-agent
Items from
Generated
Refresh
One fixed set
Last in a system card
November 2026
Last model added
November 2026
2
System cards report it, from 1 lab

No scores here yet

No lab, leaderboard or Epoch AI run has reported a score for Reasoning extraction red-teaming yet.

Labs report it only in system cards

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark. Some system cards do, below.

2 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Nov 17, 2026 GPT-6.2 Sol system card Adaptive red-teaming attack success across six attacker models, no overall score. Astra prior-turn 0.2% and same-turn 0.9%. Sol prior-turn 0.5% and same-turn 0.8% (Table 12) Robustness
Nov 17, 2026 GPT-6.2 Astra system card Adaptive red-teaming attack success across six attacker models, no overall score. Astra prior-turn 0.2% and same-turn 0.9%. Sol prior-turn 0.5% and same-turn 0.8% (Table 12) Robustness

More in Safety and alignment

All 636 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior. Inkling 2 Mini 8.30%
  • Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions. Inkling 2 Small 4.70%
  • Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior. Inkling 2 Mini 9.80%

Data audit

4 of 5 checks pass, 1 warning. Look first at the Independent runs check. No scores from any source yet. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 1 pass 3 notes
  • Warning Independent runs No scores from any source yet.
  • Note Models scored No models scored yet.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Note Frontier models No models scored yet.
  • Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 1 note
  • Note Importer No public leaderboard to read.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check