META AI Copyright claims, refereeing others to bring judge’s lawsuits

Meta On Wednesday, the company won a group of 13 authors in a large copyright lawsuit in which the company’s Çlama artificial intelligence was attended, but the judge was restricted.

The United States is protected by the doctrine of the US copyright law, which is used by the United States to train the company to train great language models or LLMs in the company.

Lawyers representing the bidders, including Coates of Sarah Silverman, because the company broke the copyright law, because the company asked the authors to use the books from the company’s books for the company’s AI model and other claims.

It should be noted that “Chamber” is generally unauthorized copy of protected works, but in this case, the bidders could not present a bidder to create a mandatory argument. Chabria wrote two defective evidence for bidders.

“In this record, the meta defeated a half-hearted argument, which is a copy of the bidders or threatens significant market losses,” he said. “This result can be a significant tension with reality.”

The experience of the meta “Copying works for a transformative purpose” is protected by a doctrine of fair use, wrote the judge.

“We appreciate today’s decision from the court,” he said. “Transformation updates, productivity and creativity for open source AI models, individuals and companies are a vital legal base to create this transformative technology for the fair use of copyright.”

Although the metadan data education experience has a negative impact on the book market, the bidders wrote that they did not need their work at the required level and wrote the judge.

The lawyers representing the plaintiffs said the decision “did not agree with respect.”

“The court fed the EU companies, without authorization of copyright owners or without paying them without paying them.

Again, the concept of “evil dispersed” in the defense, including the “public interest”, including “the understanding of the” understanding “by using copyrighted by copyright rights,” he said.

“META will suspend the development of LLM and other generative AI technologies on the roads of such a judge,” Cabria is writing. “This is nonsense.”

The judge came out of the door by saying, “The results of the consequences of the crop of the judgment,” the copyright claims associated with the EU are “limited in the great scheme of the work.”

“This is not a class action, so the judge only affects the rights of these thirds – the employees are not the number of innumerable others used to develop methane models.” “And now it should be clear, this decision is not preparing to show that the use of the copyrighted materials to cultivate its language models.”

In addition, Chabria said that the plaintiffs, which are still waiting, the bidders, which are also illegally distributed in the meter’s works illegally distributed. ”

At the beginning of this week a The federal judge ruled this AnthropicalUsing books to train You have a model The club was also “transformative” and thus satisfied the doctrine of fair use. Again, the judge said that millions of pirate books were not encountered in court in connection with the allegations of millions of pirate books to bring up anthropic and AI systems. “

“He then purchased a copy of an anthropic book, the Internet will not be responsible for theft of theft, but it can affect the degree of legal damage,” he said.

Divide: Meta Representatives rely on WhatsApp prohibitions on the facilities used by the house.