Explains segmented data quality reporting
A whole-table rate can hide a source, load or region failing completely.
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 16% | Not shown |
| gpt-6-luna | Low cost | 91% | Mastered |
| claude-haiku-5-5 | Low cost | 100% | Mastered |
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: A whole-table rate can hide a source, load or region failing completely
How it is assessed: Chooses the segments to report for sample checks
Related knobits
Builds on
Needed for
- Q-325 Recognises an aggregate hiding a total failure
- Q-406 Recognises one quality ratio over differing populations
All 30 knobits of Derive data quality rules.
Source: Data-management practice used by the stage 3 reference and review checklist; formerly data-modelling K-757