Open Models Almanac
Methodsevaluationhuman & real-world evaluation

single mention · evaluation

Dangerous-capability uplift assessment

Assesses knowledge and procedural uplift in CBRN and cyber capabilities through internal and external testing.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See human & real-world evaluation for the methods filed beside it.

Filed under evaluation › human & real-world evaluation