A group of teachers specializing in copyright law submitted a short amicus In support of the authors practicing META for the alleged training of Llama AI models in without permission.
Short, filed on Friday at the US District Court for the Northern District of California, the section of San Francisco, calls on META’s defense for fair use “a request for greater legal privileges than the courts have ever granted human writers”.
“The use of copyright -protected works for genetic models is not ‘transformative’, because the use of works for this purpose is not significantly different from their use to train human writers, which is a major original purpose of all [authors’] Works, “says the short.” The use of training is also not “transformative” because its purpose is to allow the creation of projects that compete with copywriting projects in the same scope markets that, when pursued by a speculative company such as META, also undoubtedly use “commercial”.
The International Association of Scientific, Technical and Medical Publishers, the World Trade Association for Academic and Professional Publishers, also submitted a short amicus to support the authors on Friday. The same thing happened with the copyright allianceA non -profit organization representing artists in a wide range of copyright branches, And the Union of American Publishers.
A few hours after publishing this track, a Meta spokesman pointed TechCrunch to AMICUS Flays filed by a smaller group of law teachers and the Frontier Foundation last week stake the legal position of the technological giant.
In the case, Kadrey by Meta, authors, such as Richard Kadrey, Sarah Silverman and Ta-Nehisi Coates, claimed that Meta had violated copyright using their electronic books to train models and that the company had removed copyright information from these electronic books to hide the supplies. Meta, meanwhile, claimed not only that her education is characterized as a fair use, but that the case should be rejected because the authors do not have their attitude to sue.
Earlier this month, US judge Vince Chhabria allowed the case to proceed, though it rejected part of it. In his decision, Chhabria wrote that the claim of copyright infringement is “obviously a particular injury sufficient for attitude” and that the authors have also “sufficiently claimed that Meta was deliberately removed CMI [copyright management information] to hide the copyright breach. ”
The courts weigh several copyright lawsuits AI at the moment, including the New York Times pipeline against Openai.
Updated 8:36 PM Pacific: Added references to additives.
