Proposes a DPIA with Art. 22 safeguards for automated decisions
A DPIA with human review, explanation and a contest route, for the DPO's advice
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 | 87% | Not shown |
| gemini-3.5-flash-lite | Low cost | 87% | Not shown |
| gpt-6-luna | Low cost | 87% | Not shown |
| claude-haiku-5-5 | Low cost | 87% | 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: A DPIA with human review, explanation and a contest route, for the DPO's advice
How it is assessed: The DPIA covers the decision's risks and the Art. 22(3) safeguards
Related knobits
Builds on
All 19 knobits of Record processing activities and assess privacy impact.
Source: Graded on its own by dsec-privacy