Class Action vs. Open AI 'Just the Beginning of Our Battle,' Says Authors Guild President
Without the copyrighted works of 17 plaintiff authors and a proposed class, OpenAI would have a “vastly different commercial product,” said Rachel Geman of Lieff Cabraser in a Wednesday news release announcing a copyright infringement suit (docket 1:23-cv-08292) brought by 17 authors and the Authors Guild against OpenAI in U.S. District Court for Southern New York in Manhattan. OpenAI copied authors’ works to train their large language models (LLMs) without offering choice or compensation, which “threatens the role and livelihood of writers as a whole,” said Geman, the suit's co-counsel.
Sign up for a free preview to unlock the rest of this article
Export Compliance Daily combines U.S. export control news, foreign border import regulation and policy developments into a single daily information service that reliably informs its trade professional readers about important current issues affecting their operations.
Authors Guild President Maya Shanbhag Lang, one of the plaintiffs, said the suit is “merely the beginning of our battle to defend authors from theft by OpenAI and other generative AI.” Generative AI threatens to “decimate the author profession,” said the Guild, saying the median full-time author income in 2022 was “barely over $20,000,” counting book and other author-related activities.
Authors Guild CEO Mary Rasenberger said AI threatens literary culture. “To preserve our literature, authors must have the ability to control if and how their works are used by generative AI,” she said. GPT models and other current generative AI machines “can only generate material that is derivative of what came before it,” she said: “They copy sentence structure, voice, storytelling, and context from books and other ingested texts.” AI outputs are “mere remixes without the addition of any human voice,” and “regurgitated culture is no replacement for human art.”
Plaintiff and author Jonathan Franzen called generative AI “a vast new field for Silicon Valley’s longstanding exploitation of content providers.” Authors should have the right “to decide when their works are used to ‘train’ AI,” Franzen said. If authors choose to opt in, “they should be appropriately compensated.”
OpenAI copied the plaintiffs’ works “wholesale,” without permission or consideration, and then fed their works into their LLMs “designed to output human-seeming text responses to users’ prompts and queries,” said the complaint. The LLMs allow anyone to generate automatically, either freely or “very cheaply,” texts they “would otherwise pay writers to create,” it said. Also, LLMs can “spit out derivative works: material that is based on, mimics, summarizes, or paraphrases Plaintiffs’ works, and harms the market for them,” it said.
The defendants could have trained their LLMs on works in the public domain, and they could have paid a “reasonable licensing fee” to use copyrighted works, said the complaint. What OpenAI “could not do was evade the Copyright Act altogether to power their lucrative commercial endeavor, taking whatever datasets of relatively recent books they could get their hands on without authorization,” it said.
“Unfairly and perversely,” without the plaintiffs’ copyrighted works used to train their LLMs, OpenAI would have no commercial product “with which to damage -- if not usurp -- the market for these professional authors’ works,” said the complaint. The “willful copying” makes the authors’ works “engines of their own destruction,” it said.
The complaint cited comments from OpenAI CEO Sam Altman before Congress, saying the company doesn’t want to replace creators, and creators deserve control over how their creations are used. Altman also indicated OpenAI licensed content directly from content owners for training purposes, the complaint said. “Not so from Plaintiffs,” it said. For them, OpenAI and Altman “proved unwilling to turn these words into actions.”
The plaintiffs claim direct, vicarious and contributory copyright infringement. They seek damages for “the lost opportunity to license their works, and for the market usurpation” Open AI was enabled by making the authors “unwilling accomplices in their own replacement.” They also seek a permanent injunction to prevent the harms from recurring.
In addition to Lang and Franzen, the named plaintiffs are authors David Baldacci, Mary Bly, Michael Connelly, Sylvia Day, John Grisham, Elin Hilderbrand, Christina Baker Kline, Victor LaValle, George R.R. Martin, Jodi Picoult, Douglas Preston, Roxana Robinson, George Saunders, Scott Turow and Rachel Vail. OpenAI didn't comment.