Active SDNY · consolidated

Authors Guild v. OpenAI

The consolidated SDNY author-class action against OpenAI. October 2025 brought the first U.S. ruling holding that an LLM's plot summaries of copyrighted novels may themselves constitute infringement.

🏛️ S.D.N.Y.· Lead case 1:23-cv-08292· Filed Sep 2023

Case summary

Plaintiffs
The Authors Guild · George R.R. Martin · John Grisham · Jodi Picoult · Michael Connelly · Jonathan Franzen · David Baldacci · and others (consolidated)
Defendant
OpenAI, Inc. and OpenAI affiliates; Microsoft Corp. (co-defendant in related actions)
Court
U.S. District Court, Southern District of New York
Lead case number
1:23-cv-08292
Filed
September 2023; consolidated 2024
Claim types
Direct copyright infringement · contributory infringement · vicarious liability · DMCA §1202
Status
Discovery; summary-judgment briefing underway
Class counsel
Susman Godfrey LLP and co-counsel

TL;DR

Sixteen prominent novelists and the Authors Guild sued OpenAI in late 2023, alleging GPT models were trained on their books and that ChatGPT can produce plot-level summaries close enough to the underlying works to substitute for the originals. The cases were consolidated in 2024.

In October 2025 the court denied OpenAI's motion to dismiss the output-infringement theory, holding that short, plot-level summaries of plaintiffs' novels — even where they do not quote verbatim — may infringe if they capture protected expression at a non-trivial level of detail. That ruling realigned the litigation: training-corpus arguments are no longer the only liability theory.

Why it matters

  • Outputs are now actionable, not just inputs. Until October 2025, U.S. AI cases focused on training. The Authors Guild ruling makes ChatGPT's responses themselves a copyright question.
  • Sets up parallel exposure to Bartz. If outputs alone can infringe, every LLM operator faces a separate per-output liability theory in addition to per-work training claims.
  • Discovery will be enormous. The court's order expands the universe of relevant evidence to include sample outputs across the named plaintiffs' titles, echoing the SDNY log-preservation regime in NYT v. OpenAI.
  • Plaintiffs' bar will replicate. Expect mirror complaints from non-fiction authors, screenwriters, and journalists.

More to come

Full docket timeline, the October 2025 opinion analysis, and the discovery posture for the consolidated cases coming soon. Sign up for ruling alerts to be notified when the summary-judgment opinion lands.