AI Policy Wiki
Dashboard

Tremblay v. OpenAI

medium confidence · updated 2026-08-02

Authors' class action against OpenAI over book-training data (N.D. Cal., filed June 28, 2023) — the case that absorbed the Silverman and Kadrey author claims against OpenAI, transferred by the JPML to S.D.N.Y. in April 2025 and consolidated into MDL 1:25-md-03143.

Tremblay v. OpenAI, Inc., No. 3:23-cv-03223 (N.D. Cal.), is a putative class action brought by book authors against OpenAI alleging that its models were trained on their copyrighted books without licence. Filed on June 28, 2023 by novelists Paul Tremblay and Mona Awad, it became the lead vehicle for the author claims against OpenAI in the Northern District of California, absorbing the separately filed Silverman and Kadrey actions. On April 21, 2025 the Judicial Panel on Multidistrict Litigation transferred it to the Southern District of New York under 28 U.S.C. § 1407, where it was reopened as 1:25-cv-03482 and folded into MDL 1:25-md-03143, In re: OpenAI, Inc. Copyright Infringement Litigation (Source: courtlistener.com).

Infobox

FieldValue
Original case number3:23-cv-03223 (N.D. Cal.), also docketed 4:23-cv-03223
FiledJune 28, 2023
Original plaintiffsPaul Tremblay, Mona Awad
Later named plaintiffsSarah Silverman, Richard Kadrey, Christopher Golden, Ta-Nehisi Coates, Junot Díaz, Andrew Sean Greer, David Henry Hwang, Matthew Klam, Laura Lippman, Rachel Louise Snyder, Jacqueline Woodson
DefendantsOpenAI, Inc. and six affiliated OpenAI entities; Microsoft Corporation
Original judgeAraceli Martínez-Olguín (N.D. Cal.)
TransferredApril 21, 2025, by JPML order under 28 U.S.C. § 1407
New case number1:25-cv-03482 (S.D.N.Y.), opened April 28, 2025
Consolidated intoMDL 1:25-md-03143, In re: OpenAI, Inc. Copyright Infringement Litigation

Claims

The complaint alleges that OpenAI trained GPT-series models on the plaintiffs' books, obtained from sources the plaintiffs characterise as unlicensed, and that the models can produce accurate summaries of those books as evidence that the works were ingested. The theories advanced included direct copyright infringement, vicarious infringement, DMCA § 1202(b) removal of copyright-management information, and state-law unfair-competition and unjust-enrichment claims. Microsoft was named as a defendant alongside the OpenAI entities.

Procedural history

The case was filed in the Northern District of California on June 28, 2023 with a jury demand. Sarah Silverman, Richard Kadrey and Christopher Golden filed a closely related action shortly afterwards, and the author claims were administratively brought together before Judge Araceli Martínez-Olguín, producing the consolidated plaintiff list above. As with the parallel author litigation against other developers, the court narrowed the pleadings, and the case proceeded into discovery, motions to compel, and disputes over leave to file a second amended complaint through early 2025.

On April 8, 2025 the court vacated a scheduled April 17 hearing in light of the impending transfer. The JPML transfer order issued April 21, 2025, and the case opened in the Southern District of New York on April 28, 2025 as 1:25-cv-03482 (Source: courtlistener.com). Subsequent substantive activity — including the January 2026 order compelling production of 20 million anonymised ChatGPT logs — has proceeded within the consolidated MDL rather than on the original docket (NYT v. Microsoft, OpenAI et al.).

Current status

Consolidated and pending in the S.D.N.Y. multidistrict litigation. No merits ruling on fair use has issued in the author track. The case's significance is structural as much as doctrinal: it is the point at which the scattered 2023 author suits against OpenAI became a single coordinated proceeding, which is the posture in which the fair-use question against OpenAI is now being litigated.

Relationships