Silverman is suing the tech companies over claims that ChatGPT and Meta’s AI chatbot LLaMA were trained on illegally-acquired datasets that contained a book that Silverman wrote.
Two other authors - Christopher Golden and Richard Kadrey – are also suing the companies who the suit alleges used “shadow library” websites like Bibliotik, Library Genesis, Z-Library, and others, to train their AI systems.
The Verge reported that as part of the suit the three authors claim that when prompted, ChatGPT will summarize their books, which they say is copyright infringement.
The plaintiff’s evidence includes an exhibit showing Silverman’s memoir Bedwetter being summarized by ChatGPT. The claim says the chatbot never bothered to “reproduce any of the copyright management information Plaintiffs included with their published works.”
The lawsuit against Meta alleges that the authors’ books were accessible in datasets Meta used to train its LLaMA models, one of four open-source AI Models Meta introduced earlier this year.
The lawsuit alleges that the datasets have illegal origins which can be traced back to “a copy of the contents of the Bibliotik private tracker.” The lawsuit called Bibliotik and the other “shadow libraries” listed “flagrantly illegal.”
According to The Verge, neither Silverman or either of the other two authors have commented on the case.