Propagates classification to derived columns
Every column derived from a sensitive source at its source's level and categories, or under a named rule
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 | 81% | Not shown |
| gemini-3.5-flash-lite | Low cost | 60% | Not shown |
| gpt-6-luna | Low cost | 84% | Not shown |
| claude-haiku-5-5 | Low cost | 82% | 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: Every column derived from a sensitive source at its source's level and categories, or under a named rule
How it is assessed: No derived copy of the PAN or of a direct identifier sits below its source without a rule
Related knobits
Builds on
All 16 knobits of Classify data.
Source: Graded on its own by dsec-classification