Recognises a field meaning inferred from its name
A field's meaning taken from its name with no owner confirmation.
Can AI do this?
AI does this partly Some of the tested AI models master this; others do not show it. 2 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 98% | Mastered |
| gemini-3.5-flash-lite | Low cost | 64% | Not shown |
| gpt-6-luna | Low cost | 71% | Not shown |
| claude-haiku-5-5 | Low cost | 98% | 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: The field and the confirmation needed
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: FP-007 LN_AMT "loan amount" - original or outstanding? Acceptable: Owner confirms: original principal
Also known as: Meaning inferred from the name
Related knobits
Builds on
All 13 knobits of Analyse and profile the sources.
Source: Review checklist 01-source-analysis, row 7 (DMBOK ch. 8)