About MASK
Definition and scoring
- Organisation
- Center for AI Safety
- Category
- Research
- Version
- Not evaluated
- Direction
- higher is better
- Ranking use
- Reference
- Contamination risk
- Low
Model honesty evaluation measuring whether models fabricate answers when uncertain versus appropriately abstaining. Every genuine source row stays tied to its exact model label, configuration, benchmark version and evaluation system. The summary chart shows one best compatible score per canonical product; Score 2.0 admits only explicitly mapped, frozen protocols and keeps incompatible configurations separate.
Open benchmark source ↗