Explains testing a root cause against the evidence
A cause must predict where the failure occurs and where it does not; a cause claimed for several failures must produce each by the same mechanism, and unrelated failures get causes of their own (ISO 9001:2015 clause 10.2.1).
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 95% | Not shown |
| 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: A cause must predict where the failure occurs and where it does not; a cause claimed for several failures must produce each by the same mechanism, and unrelated failures get causes of their own (ISO 9001:2015 clause 10.2.1).
How it is assessed: Answers a written question on the root cause evidence correctly
Related knobits
Builds on
Needed for
- Q-503 Recognises one root cause claimed for unrelated failures
- Q-504 Recognises a root cause the failure distribution contradicts
All 9 knobits of Analyse the root causes.
Source: Practice the stage 5 reference and review checklist rest on