Rewrites a recommendation as a verifiable requirement
A shall-requirement with a threshold and a verification method, proposed to its owner
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 | 65% | Not shown |
| gemini-3.5-flash-lite | Low cost | 33% | Not shown |
| gpt-6-luna | Low cost | 64% | Not shown |
| claude-haiku-5-5 | Low cost | 67% | 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 shall-requirement with a threshold and a verification method, proposed to its owner
How it is assessed: The requirement uses shall, can be checked by a second person, and any new threshold is marked proposed
Related knobits
Builds on
- G-352 Explains the verbal forms of requirements
- G-353 Explains how compliance with a standard is verified
All 26 knobits of Write governance policies, standards and procedures.
Source: Graded on its own by dg-policies-standards