Gracenote, Nielsen's entertainment-metadata arm, has taken OpenAI to Manhattan federal court, accusing the ChatGPT maker of quietly copying decades of its proprietary TV and movie program data to build large language models. In a complaint that reads like a programmer's worst-case scenario, Gracenote claims OpenAI reproduced its program identifiers and narrative descriptions for well known series almost word for word, threatening the core licensing business that keeps its metadata machine running. The company is asking for money damages and a court order that would bar OpenAI from using its data.
What The Suit Says
Filed in the Southern District of New York, the complaint argues that OpenAI did not just vacuum up stray facts. Gracenote says the company copied the relational scaffolding that makes its Programs Database valuable in the first place, including the sequencing and linking that tie titles, episodes and related content together for customers and commercial partners. According to Axios, the filing shows examples where ChatGPT, with only light prompting, generates program identifiers and descriptions for shows such as "Breaking Bad" and "Game of Thrones" that closely track Gracenote's own entries. The outlet notes that the suit was filed by law firm Susman Godfrey on Gracenote's behalf.
Gracenote's Claims And Damages
Gracenote says its Programs Database is registered with the U.S. Copyright Office and is handcrafted by an army of human editors whose work underpins search, discovery, recommendations and licensing deals across the entertainment industry. The company told reporters that more than 1,000 editors maintain the database and warned that OpenAI's alleged copying could erode its licensing revenue, according to Reuters. In the lawsuit, Gracenote asks for statutory and actual damages, along with a court order that would block OpenAI from using the disputed data.
Gracenote's Response
Jared Grusd, Gracenote's CEO, tried to strike a careful line between innovation and appropriation. In a message posted on Gracenote's website, he wrote that "Being pro-AI and anti-theft aren't contradictory; they are the only sustainable path forward." The post says Gracenote repeatedly approached OpenAI to talk about licensing its data before resorting to a lawsuit and frames the case as a move to protect the value created by the company's editorial teams.
OpenAI's Position
OpenAI, for its part, is sticking with a familiar defense. A company spokesperson told reporters that its models "empower innovation" and are trained on publicly available data and "grounded in fair use," comments cited by Axios. That fair-use argument, centered on the idea that training on public information is legally protected, has been a recurring theme in earlier showdowns between AI developers and content owners.
Legal Stakes And Precedent
Legal observers say Gracenote is pressing a relatively novel theory that could matter far beyond TV listings. The company is not only pointing to allegedly copied text snippets, it is claiming that the database's arrangement, sequencing and relational mapping are themselves protected. Bloomberg Law noted that the complaint goes after both individual descriptors and the broader organizing structure, a strategy that, if accepted by a judge, could expand how copyright law treats curated commercial datasets.
Where This Fits In The Bigger Picture
Gracenote's case lands in the middle of a growing pile of lawsuits over how AI systems are trained, as publishers and other rights holders look to either rein in or cash in on the use of their work. The New York Times' lawsuit against OpenAI and Microsoft, for example, raises overlapping fair-use questions, according to reporting by the AP, and courts have only begun to sketch out the rules for large-scale model training.
What To Watch Next
In the near term, expect a flurry of legal maneuvering. OpenAI is likely to move to dismiss or pare back the claims, while Gracenote will push for discovery into how the models were trained and what data went in. The Southern District of New York docket will reveal whether either side seeks an emergency injunction or quietly edges toward settlement talks. However it plays out, the result is almost certain to influence how data vendors and AI companies negotiate licensing in the next round of deals.
