Mark Zuckerberg, CEO of Meta, has hired three researchers from OpenAI to bolster his company’s work on artificial intelligence. This strategic move aims to guide Meta Platforms through its current AI challenges. The social-media corporation brought aboard Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai, all of whom previously held positions at OpenAI’s branch in Zurich, as per sources privy to this information. These individuals established OpenAI’s location in Zurich towards the end of last year. Prior to their time at OpenAI, they collaborated closely within Google DeepMind, which is the organization’s artificial intelligence division. A representative from OpenAI verified that the three scientists have departed from the organization. Zuckerberg has been on an intense recruiting drive To correct the company’s AI initiatives following the underwhelming reception of its most recent model, he has sometimes proposed giving researchers up to $100 million to become part o...
On Wednesday, a federal judge dismissed a lawsuit filed by 13 authors, such as Sarah Silverman and Junot Díaz, who alleged that Meta had infringed their copyright by using their books to train their artificial intelligence system.
Judge Vincent Chhabria concluded that Meta had relied on "fair use" when it utilized a dataset comprising approximately 200,000 books—including those belonging to the plaintiffs—to train its Llama language model. This ruling came after similar ruling Issued on Monday in a lawsuit against Anthropic regarding its language model, Claude.
“We value the court’s decision made today,” stated a representative from Meta. “Open-source AI models are driving significant advancements, enhancing efficiency, and fostering innovation for both individuals and businesses. The responsible usage of copyrighted materials under fair use provisions is crucial for developing these groundbreaking technologies.”
Chhabria dismissed the plaintiffs' assertion that the corporation was involved in "unrestricted piracy" with the creation of the model. The judge determined that Llama is incapable of producing reproductions exceeding 50 words, deeming the AI model as "transformative."
He was more receptive to the idea that AI might obliterate the market for genuine creations by producing countless inexpensive imitations. He noted this probably wouldn’t qualify as “fair use,” even if the output differed from the source material.
What matters most to copyright law is maintaining the motivation for people to create.
Artistic and scientific creations," the judge stated. He further explained that fair use generally does not cover copying that substantially reduces the potential income for copyright owners, thereby considerably undermining the motivation to produce new work in the future.
However, Chhabria determined that the authors in question failed to demonstrate a reduction in their book sales, or that Llama is expected to cause such an impact.
Meta provided proof that its copying has not led to market damage," he noted. "The plaintiffs did not present any empirical evidence to dispute this... The only thing the plaintiffs offered was conjecture.
Owners of copyrights have filed numerous lawsuits against artificial intelligence firms, claiming that using their protected material for training purposes without permission is unlawful.
Chhabria emphasized that his decision is based solely on the information presented to him, and the result might vary significantly in other instances.
"Within the broader perspective, the impacts of this decision are confined," he penned.
Boies Schiller Flexner LLP, who acted for the plaintiffs, stated they disagree with the decision.
The court stated that AI firms which “use protected content in their systems without obtaining consent or making payments to the copyright owners” are typically breaking the law, according to a representative from the company. However, even though Meta has an unrivaled history of unauthorized use of copyrighted materials, the court sided with Meta. We do not concur with this decision, we respectfully dissent.
The law firm's representatives chose not to comment on their intentions regarding an appeal.
The authors who brought the case Are Silverman, Díaz, Richard Kadrey, Christopher Golden, Ta-Nehisi Coates, Andrew Sean Greer, David Henry Hwang, Matthew Klam, Laura Lippman, Rachel Louise Snyder, Jacqueline Woodson, Lysa TerKeurst, and Christopher Farnsworth.
The lawsuit alleges that Meta used "shadow libraries" to obtain millions of copies of pirated books. According to the filings, Meta's engineers used BitTorrent to download the large volume of data, which involves downloading data from multiple sources and, in some cases, reuploading it.
The lawsuit claimed that Meta infringed upon the authors’ copyrights during both the AI training phase and when downloading and redistributing unauthorized libraries. Although the judge dismissed the argument related to AI training, he did not address the matter concerning torrents, leaving this aspect undecided.
More from
- Google's Sanjay Gupta Sees AI as 'Magic Wand' for Asia's Storytellers at APOS
- Xbox Collaborates with Meta Quest for Exclusive VR Headset Release