Recognises a manual step outside the control matrix
A step in the real production chain where data is changed by hand, with no control over it. The classic BCBS 239 finding.
Can AI do this?
Not settled yet The benchmark tested this, but the models' answers do not settle yet whether they master it. 0 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 70% | Not shown |
| gemini-3.5-flash-lite | Low cost | 30% | Not shown |
| gpt-6-luna | Low cost | 29% | Not shown |
| claude-haiku-5-5 | Low cost | 71% | Not shown |
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: The production step and the missing control
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: report_production step 5: "Analyst applies management overlay adjustments in a spreadsheet before the PDF is produced" — no control Acceptable: A control: adjustments logged with reason and amount, approved by the CRO, reconciled to the published figures; or the step removed
Also known as: Manual step not recorded as a control
Related knobits
Builds on
Usually followed by
All 21 knobits of Map BCBS 239 to controls.
Source: Review checklist 04-regulatory-mapping, row 8 (BCBS P3, P7; ECB)