Recognises a definition failing the substitution test
Replacing the term with its definition in a normal sentence changes or breaks the meaning.
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-3.1-pro-preview* | Flagship | 100% | Mastered |
| 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 | 81% | Not shown |
| gpt-6-luna | Low cost | 82% | Not shown |
| claude-haiku-5-5 | Low cost | 97% | Mastered |
* From the earlier benchmark of 2026-10: 10 data-modelling test tasks, 3 runs per task, graded by claude-opus-5-5. These models have not taken the current 123 tasks yet; the two methods differ, so compare their figures with care. They are left out of the “most models” counts and the best-model picks.
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 sentence where substitution breaks and why
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: arrears rate: loans that are overdue. "The arrears rate rose to 3 %" becomes "The loans that are overdue rose to 3 %" Acceptable: arrears rate: proportion of loans whose arrears day count is above zero at the end of a given period. "The proportion of loans whose ... rose to 3 %" still reads correctly
Also known as: Fails substitution (business term)
Related knobits
Builds on
All 18 knobits of Define business terms in the glossary.
Source: Review checklist 03-business-glossary (metadata-management), row 12; formerly data-modelling K-212