Federal judge sides with Meta in lawsuit over training AI models on copyrighted books

A federal judge ranked on the side of Meta on Wednesday in a trial against the company by 13 perpetrators of books, whose Sarah Silverman, who alleged that the company had illegally formed its AI models on their works protected by copyright.

Federal judge Vince Chhabria published a Summary judgment – which means that the judge was able to decide the case without sending him to a jury – in favor of Meta, noting that the training of the company of the models of AI on the books protected by copyright in this case was the doctrine of the “remuneration” of the law on copyright and was therefore legal.

The decision comes only a few days after a The federal judge has stored on the side of the anthropogical in a similar trial. Together, these cases promise to be a victory for the technological industry, which has spent years in legal battles with media companies arguing that the formation of AI models on works protected by copyright is fair use.

However, these decisions are not the radical victories that some companies hoped – the two judges noted that their affairs were limited.

Judge Chhabria clearly indicated that this decision does not mean that any training on AI models on works protected by copyright is legal, but rather that complainants in this case “have made the wrong arguments” and have not developed sufficient evidence to the support of the good.

“This decision does not represent the proposal according to which the use by META of documents protected by copyright to form its linguistic models is legal,” said Judge Chhabria in his decision. Later, he said: “In cases involving uses like those of Meta, it seems that complainants will often win, at least where these cases have better developed files on the effects of the defendant’s use market.”

Judge Chhabria judged that Meta use of works protected by copyright in this case was a transformative – which means that the company’s models of the company did not reproduce the books of the authors.

In addition, the complainants did not convince the judge that the copy of the books by Meta harmed the market for these authors, which is a key factor to determine whether the copyright law has been raped.

“The complainants did not present any significant evidence on the dilution of the market,” said judge Chhabria.

Anthropic and Meta victories involve the training of AI models on books, but there are several other active proceedings against technological companies for the formation of AI models on other works protected by copyright. For example, The New York Times continues Openai and Microsoft for the training of AI models on press articles, while Disney and Universal continue the middle of the middle For the training of AI models on movies and television shows.

Judge Chhabria noted in his decision that the defenses for fair use greatly depend on the details of a case, and certain industries may have stronger arguments of fair use than others.

“It seems that the markets for certain types of works (such as press articles) could be even more vulnerable to indirect competition from AI results,” said Chhabria.