Comedian Sarah Silverman has filed lawsuits against Meta and OpenAI, the parent company of ChatGPT.
The filing claims that the two companies have broken copyright law by using protected content to train artificial intelligence models, Reuters reported.
Programs such as ChatGPT, Meta’s LLaMA and others use text, images, videos and other data to learn how to respond to a user’s prompt, Fox Business and The Verge reported.
Silverman is joined by Richard Kadrey and Christopher Golden in the proposed class action lawsuits filed in San Francisco federal court on Friday.
Silverman’s book “Bedwetter,” Kadrey’s book “Sandman Slim” and Golden’s book “Ararat” are listed in the suit, according to The Verge. The authors said that ChatGPT did not “reproduce any of the copyright management information Plaintiffs included with their published works.”
In the allegations against OpenAI, the suit claims that the platform summarizes the writers’ work.
“The summaries get some details wrong,” but shows that it “retains knowledge of particular works in the training dataset,” the lawsuit alleges, according to Reuters.
The suit also alleged that the writings were in “shadow library” websites such as Bibliotik, Library Genesis and others used by Meta to train LLaMA, The Verge reported.
Both lawsuits say the writers did not consent for either company to use their copyrighted materials. The suits have six counts of alleged violation.
Meta and OpenAI did not respond to Reuters’ or Fox Business’ request for comment.
The lawsuits seek monetary damages on behalf of copyright owners whose works were allegedly used.
This isn’t the first lawsuit against an AI company. Getty Images also has filed a lawsuit against Stability AI claiming that it trained its Stable Diffusion program on “millions of images protected by copyright,” The Verge reported.
AI Lawsuit by National Content Desk on Scribd
