Recognises a data quality service level threshold with no rationale
A threshold not derived from the use.
Can AI do this?
People still lead Few or none of the tested AI models showed they master this. 0 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 82% | Not shown |
| gemini-3.5-flash-lite | Low cost | 5% | Not shown |
| gpt-6-luna | Low cost | 5% | Not shown |
| claude-haiku-5-5 | Low cost | 5% | 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 threshold derived from the tolerance of the use
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: SL-003 ledger balance vs GL 95%, "industry standard" Acceptable: 99.9%, set from the LCR and DGS file tolerances the balance feeds
Also known as: Threshold with no rationale
Related knobits
Builds on
All 10 knobits of Set data quality service levels and the scorecard.
Source: Review checklist 07-sla-scorecard, row 2 (BCBS 239 P3; DMBOK §2.7)