Publisher Ziff Davis sues OpenAI for copyright infringement

Major Publisher Files Federal Lawsuit Against OpenAI

Digital media giant Ziff Davis has filed a lawsuit against OpenAI in a Delaware federal court, alleging widespread misuse of its publications to train artificial intelligence models. Ziff Davis claims that the OpenAI team “intentionally and relentlessly” exploited copyrighted content to develop the technology behind its popular chatbot ChatGPT[1][2].

Scope of the Lawsuit and Allegations

The lawsuit accuses OpenAI of:
  • Reproducing exact copies and creating derivatives of Ziff Davis content for training AI models
  • Infringing on Ziff Davis copyrights and diluting its trademarked brands
  • Using proprietary reporting and advice articles from flagship sites such as ZDNet, PCMag, CNET, IGN, Lifehacker, and Mashable to generate chatbot responses
With more than 45 sites globally generating an average of 292 million monthly visitors, Ziff Davis is among the largest U.S. publishers to challenge OpenAI in court over the alleged unauthorized use of its content[2].

Industry Context and Response

The complaint against OpenAI is part of a wave of copyright lawsuits from news organizations, authors, and artists contending that generative AI technologies are being developed on the backs of stolen intellectual property. The New York Times and Dow Jones are among other publishers currently pursuing similar legal action. An OpenAI spokesperson maintains that its models “empower innovation, and are trained on publicly available data and grounded in fair use.” Industry peers including Google and Meta Platforms have made similar defenses, arguing AI training falls under the fair use provision in U.S. copyright law[1]. A spokesperson for Ziff Davis declined to comment further on the ongoing litigation.

What’s Next?

The Ziff Davis suit raises the stakes in the ongoing debate over AI development, copyright law, and digital publishing rights. The outcome is likely to have far-reaching effects on how AI companies source training data and how publishers protect their intellectual property moving forward[1][2].