Links a correction to the record it corrects
The correction carries the original's key and is netted or applied against it
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 | 83% | Not shown |
| gemini-3.5-flash-lite | Low cost | 14% | Not shown |
| gpt-6-luna | Low cost | 83% | Not shown |
| claude-haiku-5-5 | Low cost | 83% | 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: The correction carries the original's key and is netted or applied against it
How it is assessed: Every reversal can be traced to its original
Related knobits
Builds on
All 15 knobits of Design the integration patterns.
Source: Graded on its own by dii-integration-patterns