Meta predicted last year that its generative AI products will generate $2 billion to 3 billion in revenue by 2025 and between $460 and $1.4 trillion in revenue by 2035. According to court documents that were unsealed on WednesdayThe documents submitted by attorneys representing book authors who are suing Meta over what they claim is the unauthorized training of its AI on their works do not indicate what Meta considers a “generative AI” product. But it’s well known that the tech giant earns money — and will continue to make money — from generative AI.
Meta is in revenue-sharing agreements for certain companies who host its open Llama models. The company launched an API to customize and evaluate Llama models. Mark Zuckerberg, CEO of Facebook, said that Meta AI, Meta’s AI assistant, could eventually show ads and provide a subscription service with additional features. During the company’s Q1 earnings conference call on Wednesday, said.
Court documents also reveal that Meta spends a huge amount on its AI product group. The documents reveal that the company’s “GenAI budget” in 2024 was more than $900 million and that this year it could be over $1 billion. This does not include the infrastructure required to run and train AI model. Meta had previously stated that it planned to spend $60 billion-$80 billion in capital expenditures by 2025, primarily for new data centers. These budgets could have been higher if they had included deals to license the books from the authors who are suing Meta. Meta, for example, discussed spending up to $200 million in 2023 to acquire training data. Of that, approximately $100 million would have gone towards books, according documents. The company decided to go with another option: massive ebook piracy.June 5
Meta has developed [open] AI[transformational]models that power incredible innovation, productivity and creativity for both individuals and businesses. Fair use of copyrighted material is essential to this. We disagree with [the authors’] ‘s assertions and the full records tell a different tale. We will continue to defend ourselves and protect the development of generative AI to the benefit of everyone.”
Kyle Wiggers, TechCrunch’s AI Editor. His writings have appeared in VentureBeat, Digital Trends, and a variety of gadget blogs, including Android Police and Android Authority, Droid-Life and XDA-Developers. He lives in Manhattan, with his music therapist partner.
View Bio
