Mark Zuckerberg Mark Zuckerberg at the annual event of Meta Connect, at the Company’s Meta Connect annual event, at the Menorok, California headquarters, September 25, 2024.
Orbitoso Manuel | Reuters
Metabil On Wednesday, the company had 13 authors who took part in the company’s artificial intelligence model, but the judge was limited to this.
Thanks to the US judge of the United States, the company’s use of their own great language models or LLMs is protected by copyright Doctrine of US copyright.
Lawyers who represent applicants, including Sarah Silverman and Ta-Nechisi Kanos, have violated the law of the nation’s copyright, as the company did not seek to use the company’s books to the company AI model, and other complaints.
Chahbria Chabriya said that “protected works illegal to download” protected works, but in this case, applicants failed to provide evidence that the meta could use books for “market damage”. Chachbria wrote that applicants had two flaws for their business.
“This record meta won half of the sincere half of the applicants, which causes or threaten the market,” said Chabriya. “This statement can be considered significantly due to reality.”
The practice “Copying the work for transformation for transformation” is protected by a fairy doctrine, wrote a judge wrote.
“We appreciate the decision of the court today,” said the meta representative. “Transformation innovations, productivity and creativity, productivity and creativity for individuals and companies – important legal framework for creating transformative technologies for individuals and companies.”
Even if the meta-based data learning experience can have evidence that adversely affects the book market, the plaintiffs did not fulfill their work correctly, wrote a judge wrote.
Applicants who represent applicants did not respond to the request of the explanation.
CHABRIA metama’s defense, including the concept of “public interests”, is prohibited from using “public interests” in “public interests” if it is not prohibited from the company and other business.
“META, such power will stop the development of LLMs and other generational AI technologies in their own ways,” Chagria wrote. “It’s Nonsense”.
Judge says that open authorial cases for other authors have opened similar authoring trials against META and say that “the consequences of this power are limited in the qualified scheme.”
“This is not a classwork, so the authorities affect the rights of these thirteen authors – it does not teach their model, teaching their model,” he said. “And now it’s a decision to use copyrighted metamar-protected materials to teach its language models.”
In addition, Chagria still expected that it still has an expected claim that the meta can be obtained for illegal distribution of their works (through torrenting). »
By the beginning of this week, a Headed by a federal judge that AnthropicUse books to train it You have a model Claude also satisfies the doctrine on “transformation”, thus making a fair application. This judge still said that anthropic said, he said that he had uploaded millions of robberies to train his AI system.
“Then he bought a copy of the book before he previously stolen the stolen.
Consideration: Whatsapp is prohibited from devices used by the meta, the Chamber of Representatives.
