Gracenote Media Services v OpenAI
Gracenote Media Services, LLC v. OpenAI (Case No. 1:26-cv-01947, S.D.N.Y.) is a copyright infringement lawsuit filed by Nielsen subsidiary Gracenote against OpenAI, alleging unauthorized copying and use of Gracenote's proprietary media metadata database to train and power AI products including ChatGPT.[1][2]
Parties
- Plaintiff
- Gracenote Media Services, LLC, a Nielsen subsidiary that maintains a copyrighted entertainment metadata database (the Programs Database) covering music, video, and sports content, including relational architecture and descriptive metadata used by media distributors worldwide[1][2]
- Defendants
- OpenAI Foundation (f/k/a OpenAI, Inc.), OpenAI Group PBC, OpenAI GP, LLC, OpenAI, LLC, OpenAI OpCo, LLC, OpenAI Global, LLC, OAI Corporation, LLC, and OpenAI Holdings, LLC (collectively "OpenAI")[1]
- Plaintiff Counsel
- Susman Godfrey LLP[1]
Court
- Court
- United States District Court, Southern District of New York
- Docket Number
- 1:26-cv-01947[3]
Claims
Gracenote alleges that OpenAI willfully copied its copyrighted metadata and relational framework without permission or license to train large language models for products like ChatGPT, ChatGPT Enterprise, and developer APIs. Specific allegations include:[1][4]
- Unauthorized reproduction of Gracenote's copyrighted Programs Database and relational architecture
- OpenAI models reproduce exact copies of Gracenote data verbatim (e.g., exact "Game of Thrones" episode summaries and metadata)
- AI outputs enable competitors to build substitutive metadata products, threatening Gracenote's core business
- Willful infringement seeking actual damages, statutory damages, and injunctive relief
- Jury trial demanded
Current Status
The case is newly filed and pending. It has been affirmed as related to the Multi-District Litigation In re OpenAI Copyright Infringement Litigation. No rulings, motions, or settlements have been reported as of April 2026. The complaint has been noted as the 90th U.S. copyright lawsuit against AI companies.[5][3]
Significance
This case raises the novel issue of whether database metadata and relational architecture — not raw creative content but structured descriptive data — qualifies for copyright protection against AI training. A ruling in Gracenote's favor could significantly expand the scope of copyrighted material that AI companies must license, beyond traditional books, music, and images.[4][2]
Related Cases
- Bartz v Anthropic PBC — Authors' $1.5B settlement over pirated book training
- Kadrey v Meta Platforms Inc — Authors allege Meta trained LLaMA on pirated books
- Carreyrou v Anthropic PBC — Opt-out authors sue multiple AI companies
References
- ↑ 1.0 1.1 1.2 1.3 1.4 Gracenote Media Services v. OpenAI Complaint Overview
- ↑ 2.0 2.1 2.2 2.3 Axios, "Nielsen's Gracenote sues OpenAI for copyright infringement", March 10, 2026
- ↑ 3.0 3.1 3.2 Justia Docket: Gracenote v. OpenAI
- ↑ 4.0 4.1 VitalLaw, "Entertainment Metadata Creator Sues OpenAI"
- ↑ ChatGPTIsEatingTheWorld, "Gracenote Media v. OpenAI Case Tracker"