Editorial review · 260601-001
How ORA’s piece on The Workers Who Trained the AI Are Being Let Go With Nothing scored.
Read the article →Solid reporting. Some issues but credible overall. The reader is well-served.
Accuracy
Core legal claims about Ireland's Redundancy Payments Act, the two-year threshold, and the €600 weekly cap are correct and properly cited. The 720 workers, 400 sub-threshold figure, and Meta's 8,000 cuts with four months' severance are post-cutoff but attributed to WIRED, RTÉ, and The Journal (-3 for one minor hedge laundering on the CWU quote framing). One minor deduction for the unsourced specific claim that Covalen rebranded from Majorel Ireland (-5).
Balance
ORA engages three named counterpoints (pre-AI precarity, partial-offset annotation demand, EU threshold norms) substantively rather than as strawmen. The Irish FDI tension is acknowledged honestly rather than resolved rhetorically. Tone slants pro-worker throughout without equivalent treatment of Covalen's or Meta's operational rationale beyond the WIRED statement (-5).
Concerns (3)
- minoraccuracy
“Covalen — the outsourcing firm that employs the workers, rebranded from Majorel Ireland”
Specific corporate history asserted without citation.
Evidence: No source supports the Majorel rebrand claim in the footnotes.
- minoraccuracy
“720 workers, ~400 below threshold, Meta 8,000 cuts with four months minimum”
Post-cutoff figures, source attributed.
Evidence: Attributed to WIRED, RTÉ, The Journal, and AI Weekly; not independently verifiable here.
- minorbalance
“(overall framing)”
Tone slants pro-worker without equivalent operational framing.
Evidence: Meta and Covalen positions appear only via one short statement; no industry voice engaged at length.
Reproducibility
How this review works: read the methodology. Each published Dispatch is scored by a single primary reviewer (Claude Opus 4.7) against the public rubric. A second model (Gemini 2.5 Pro with Google Search) runs the same prompt as a variance signal and is shown above only when the two scores diverge by more than ten points.