Explains a data classification scheme
Levels (public, internal, confidential, restricted) with handling rules, categories (personal, direct identifier, special category, cardholder data, banking secrecy) with minimum levels, applied to every column (ISO/IEC 27002:2022 5.12-5.13).
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 | 99% | 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: Levels (public, internal, confidential, restricted) with handling rules, categories (personal, direct identifier, special category, cardholder data, banking secrecy) with minimum levels, applied to every column (ISO/IEC 27002:2022 5.12-5.13).
How it is assessed: Answers a written question on the classification scheme correctly
Related knobits
Needed for
- S-201 Recognises a card number not classified as cardholder data
- S-204 Recognises an unclassified column
- S-206 Recognises a level below its category's minimum
- S-210 Recognises a classification disagreeing with the physical model
- S-211 Recognises a classification level with no handling rules
- S-252 Explains classification of derived data
All 16 knobits of Classify data.
Source: Practice the stage 2 reference and review checklist rest on