On Wednesday, a federal judge supported Meta in a case against the author of six books, including Sara Silverman, who complained that the company had illegally trained its AI models in their copyright work.
Federal judge has issued Vince Churia Short verdict – That means the judge was able to decide on the case without sending any jury – in favor of Meta, the training of the AI models in the copyright books in this case was under the “fair use” of copyright law and thus legal.
The decision came just a few days after the federal judge was in favor of the ethnic party in a similar case. Together, these cases are a win for the technology industry, which has spent years of legal fighting with media agencies that argue that copyrighted work is the only use of AI models in training.
However, these decisions are not to win for some of the expected companies – both judges mentioned that their cases were limited to opportunities.
The judge made clear that this decision does not mean that all AI model training in copyrighted works is legal, but in this case the plaintiffs have “made the wrong argument” and fail to develop adequate evidence in support of the correct.
Judge Chambariya said in his decision, “The verdict does not stand in favor of the proposal that the use of copyrighted materials for training its language models is legal.” Later, he said, “The plaintiffs will often win, at least where these cases have a better and better record on the market influence of the dispute,” he said.
Judge Chambariya ruled that in this case the meta was converter to the use of copyrighted works – meaning the company’s AI models did not just reproduce the authors’ books.
Furthermore, the plaintiffs failed to convince the judge that copied the Meta books damaged the market for those authors, which is the main reason for determining whether copyright law has been violated.
“The plaintiffs have not presented any meaningful proof of the market decrease,” the judge said.
Both anthropics and meter are involved in the training of AI models in the books, but there are several active cases for training AI models in other copyrighted works against technology companies. For example, New York is suing OpenAI and Microsoft for training for AI models in New York New York News articles, while suing mid -journey for training for AI models on Disney and Universal Film and TV shows.
Judge Chambariya mentions in his decision that the defense of a fair use depends a lot on the details of a case and the logic of fair use may be stronger than others in some industries.
“Looks like a certain type of job markets (such as news articles) may be more risky in indirect competition than AI outputs,” said Churia.