The legal pressure on artificial intelligence developers continues to mount. Gracenote, the entertainment metadata firm owned by Nielsen, has filed a copyright infringement lawsuit against OpenAI in a San Francisco federal court. The core allegation is that OpenAI used Gracenote's proprietary data—and the unique structural framework organizing it—without permission or payment to train its models.

Gracenote's business involves creating detailed descriptors and identifiers for movies, music, and TV shows. This data powers content discovery for major television providers and streaming services. While many lawsuits have challenged the use of copyrighted text and images for AI training, Gracenote's complaint adds a novel argument. It claims OpenAI copied not just the factual data points, but the specific sequencing and relational architecture of its databases, which the company considers a separate creative work.

The filed complaint is pointed. "Defendants could have paid to license Gracenote Data. Or they could have trained their models solely on public domain information," it states. "They did neither." Gracenote alleges its prior outreach to OpenAI to discuss a licensing deal was ignored. This legal action contrasts with Gracenote's other recent business; the company has established formal partnerships to supply data for AI projects at Samsung and Google. The case underscores the unresolved tensions between rapid AI development and the ownership of specialized data systems that fuel it.

Source: Engadget