Recognises inconsistent terminology across the register
The same thing under different words in different requirements (payer, counterparty, account holder used interchangeably), or one word for two things. Stage 2 will either merge things that differ or split one thing in two.
Can AI do this?
AI does this well Most of the tested AI models master this. 9 of 10 models master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.1-pro-preview | Flagship | 78% | Not shown |
| gpt-6-astra | Flagship | 100% | Mastered |
| claude-fable-5-1 | Flagship | 100% | Mastered |
| gemini-3.8-flash | Mid tier | 100% | Mastered |
| gpt-6.1-sol | Mid tier | 100% | Mastered |
| claude-sonnet-5-5 | Mid tier | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 100% | Mastered |
| gpt-6-luna | Low cost | 100% | 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: The variant words and the term to agree with the steward
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: DR-004 speaks of the disputes specialist and DR-12 of payments handlers, and nothing says whether they are one role or two Acceptable: One word per thing across the register, chosen with the steward; the variants listed for stage 2 as synonyms.
Also known as: Inconsistent terminology (data requirement)
Related knobits
Builds on
All 47 knobits of Capture data requirements.
Source: Review checklist 01-requirements, row 24; 29148 5.2.6 (comprehensible)