Recognises a data principle contradicting a strategy goal
A principle that, applied, would block an approved goal — or a goal that can only be reached by breaking a principle.
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 | 60% | Not shown |
| gemini-3.5-flash-lite | Low cost | 48% | Not shown |
| gpt-6-luna | Low cost | 61% | Not shown |
| claude-haiku-5-5 | Low cost | 59% | 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 principle, the goal and the conflict
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: DP-009 Business lines may adjust reported figures to reflect management judgement against SG-001 (no manual adjustment to the credit risk report) Acceptable: Narrow the principle ("...in a documented overlay, outside the risk data mart, approved by the CRO") or change the goal, with the sponsor deciding which
Also known as: Principle that contradicts a goal
Related knobits
Builds on
- G-153 Explains the structure of a data principle
- G-154 Explains traceability from initiatives to goals to business objectives
All 25 knobits of Set the data strategy and principles.
Source: Review checklist 01-strategy-principles, row 15 (38505 strategy and conformance principles)