Which NIST AI RMF MEASURE subcategories can a third party evidence?
A third party can evidence the measurement-flavoured subcategories: MEASURE 2.1 (documented test sets, metrics, TEVV tools), 2.3 (performance under deployment-like conditions), 2.4 (in-production monitoring), 2.5 (validity, reliability, generalizability…
Answer.
A third party can evidence the measurement-flavoured subcategories: MEASURE 2.1 (documented test sets, metrics, TEVV tools), 2.3 (performance under deployment-like conditions), 2.4 (in-production monitoring), 2.5 (validity, reliability, generalizability limits), and 2.13 (effectiveness of the TEVV process itself). Planisphere maps to each via the probe set, a reproducible fielded run, the re-run cadence, a documented coverage ceiling, and a bit-exact replay verdict. It does not cover human-subject (2.2), environmental (2.12), or the privacy/fairness adjudications of 2.10/2.11.
Cite-anchor: NIST AI 100-1 (AI Risk Management Framework) · MEASURE function
The mark behind the answer.
Measure function · TEVV · 19 subcategories · the audit-triad home.
More on NIST AI RMF · MEASURE.
Try NIST AI RMF · MEASURE with a free test key.
Start with the free sandbox. Planisphere measures model behaviour and emits a reproducible, sha-pinned record — it does not certify, file, or give legal advice.