Recognises a model review planned by its modeller
The reviewer is the modeller, or no review is planned.
Can AI do this?
People still lead Few or none of the tested AI models showed they master this. 0 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 90% | Not shown |
| gemini-3.5-flash-lite | Low cost | 90% | Not shown |
| gpt-6-luna | Low cost | 90% | Not shown |
| claude-haiku-5-5 | Low cost | 10% | Not 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: An independent reviewer and the review at each release
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: roles: {modeller: "Mart K.", reviewer: "Mart K."} Acceptable: An independent reviewer; stage 6 at every release
Also known as: Review by the modeller
Related knobits
Builds on
All 10 knobits of Plan the modelling and set its standards.
Source: Review checklist 00-planning-standards, row 5 (DMBOK §6)