Explains the kinds of preventive and corrective data quality actions
Preventive: entry controls, training of data producers, rules defined and enforced, quality demanded from data suppliers, governance and formal change control. Corrective: automated correction, manually directed correction (reviewed before it is applied) and manual correction, each with its own risk.
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 | 78% | Not shown |
| gemini-3.5-flash-lite | Low cost | 69% | Not shown |
| gpt-6-luna | Low cost | 69% | Not shown |
| claude-haiku-5-5 | Low cost | 69% | 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: Preventive: entry controls, training of data producers, rules defined and enforced, quality demanded from data suppliers, governance and formal change control. Corrective: automated correction, manually directed correction (reviewed before it is applied) and manual correction, each with its own risk.
How it is assessed: Answers a CDMP-style multiple-choice question on the preventive and corrective actions
All knobits of Data quality.
Source: DAMA-DMBOK: Data Management Body of Knowledge, 2nd edition (revised), chapter 13: Data Quality, element QV-25