According to newly made public parts of a creation he gave late last year, CEO Mark Zuckerberg seems to have utilized YouTube's fight to stop illegal content to assist his company's usage of a data set containing copyrighted e-books.
The statements relate to the AI copyright case Kadrey v. Meta and were involved in a complaint that the plaintiffs' lawyers filed with the court. This case is one of many that pit AI companies against authors and other intellectual property holders as they integrate their way through the American legal system. The defendants in these lawsuits, which are AI businesses, typically maintain that training on protected content constitutes "fair use." Many copyright holders don't agree.
Zuckerberg Defends Meta's Use of Pirated Data
According to parts of a transcript posted Wednesday night, Zuckerberg claimed throughout his deposition that YouTube "may end up hosting some stuff that people copy for some time, but YouTube is trying to take that stuff down." "And I would imagine that the great majority of the content on YouTube is of a decent calibre and that they are enabled to do so."
Some hints about Zuckerberg's views on copyright content and fair usage can be discovered in excerpts from his deposition. Nevertheless, It should be underlined that the deposition's complete transcript was not created publicly.
Zuckerberg aims to justify Meta's use of LibGen, a training data collection of e-books, to create its Llama family of AI models, based on the deposition nuggets. AI companies like OpenAI have popular models competing with Meta's Llama.
After worries about the potential legal ramifications among Meta's AI management and research teams, Zuckerberg reportedly approved using LibGen to train at least one of the company's Llama models, based on court documents this week.
When questioned by David Boies, one of the plaintiffs' lawyers, Zuckerberg declared why it would be irrational to restrict the use of a data collection such as LibGen.
Given that some of the content on YouTube might be secured by copyright, would I wish to establish a policy against its use? "No," he responded. "In certain situations, imposing such a complete restriction might not be the best course of action." Zuckerberg claimed Meta should be "pretty careful about" teaching copyrighted content.
New Allegations
Since the complaint was submitted in 2023 in the United States District Court for the Northern District of California, San Francisco Division, the lawyers for the plaintiffs in the Kadrey v. Meta lawsuit have made different amendments.
New allegations against Meta include that the corporation cross-referenced some stolen books in LibGen with copyrighted novels that were provided for license according to the plaintiffs' lawyer’s most recent amended lawsuit, submitted late Wednesday. Lawyers state Meta employed this strategy to assess if maintaining a licensing deal with a publisher was wise.
According to the modified requests, Meta purportedly trained its most recent generation of Llama models, Llama 3, using LibGen. Additionally, the plaintiffs say that Meta is training its next-generation Llama 4 models utilizing the data collected.