Comedian Sarah Silverman Joins Class-Action Lawsuits Against OpenAI and Meta for Copyright Infringement
Renowned comedian Sarah Silverman has joined forces with authors Christopher Golden and Richard Kadrey to file class-action lawsuits against OpenAI and Meta. The lawsuits, filed in the San Francisco Division of the U.S. District Court of the Northern District of California, claim that the companies engaged in copyright infringement by utilizing the plaintiffs’ protected works to train their artificial intelligence (AI) programs.
According to court papers, Silverman and the authors allege that OpenAI and Meta copied and used their copyrighted materials without permission, scraping content from illicit online “shadow libraries” that house thousands of books. The lawsuit against Meta specifically references the company’s research paper on LLaMA, its large-language model used for training chatbots. The plaintiffs assert that text from shadow libraries was included within the training dataset, which contains some of their copyrighted works.
Both OpenAI and Meta have yet to respond to requests for comment regarding the lawsuits. The plaintiffs seek compensation for damages and injunctive relief, potentially requiring alterations to the LLaMA and ChatGPT programs.
The source of training datasets for OpenAI’s ChatGPT program remains unclear, but the lawsuit suggests that the program’s ability to generate summaries of the authors’ works is only possible if it was trained using their copyrighted materials. The exhibit in the lawsuit includes a generated summary by ChatGPT of Silverman’s memoir, “The Bedwetter,” which highlights her struggle with bedwetting during her formative years.
In addition to this ongoing litigation, attorneys Joseph Saveri and Matthew Butterick, who represent the three authors, are involved in separate legal action against Copilot, an AI-powered coding assistant on GitHub, and an image generator produced by Stability AI. The lawyers claim that much of the material used in the training datasets of OpenAI and Meta, including the plaintiffs’ books, was copied without consent, credit, or compensation.
These lawsuits are part of a growing trend of legal actions that aim to establish the parameters regarding how AI systems learn and the role copyright laws play in the content they use for training datasets. Experts predict that similar lawsuits will emerge as the boundaries of AI technology are further explored.
While Robert deBrauwere, a digital media and intellectual property specialist at law firm Pryor Cashman, anticipates more such cases, he is not involved in the Sarah Silverman lawsuit.