Explains principle exceptions
A departure from a principle is recorded with its reason and its owner's approval (TOGAF dispensation).
Can AI do this?
Not settled yet The benchmark tested this, but the models' answers do not settle yet whether they master it. 1 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 99% | Mastered |
| gemini-3.5-flash-lite | Low cost | 52% | Not shown |
| gpt-6-luna | Low cost | 52% | Not shown |
| claude-haiku-5-5 | Low cost | 52% | 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: A departure from a principle is recorded with its reason and its owner's approval (TOGAF dispensation).
How it is assessed: Answers a written question on the principle exception correctly
Related knobits
Builds on
Needed for
All 16 knobits of Record the target architecture and its decisions.
Source: Practice the stage 4 reference and review checklist rest on