Anthropic to Pay $1.5 Billion for AI Training Data Piracy of 500,000 Works

Anthropic Agrees to $1.5 Billion Settlement for AI Training Data Piracy In a landmark agreement, Anthropic has agreed to pay $1.5 billion to authors whose works were pirated to train its artificial intelligence models. The settlement, which covers 500,000 works, is believed to be the largest publicly reported recovery in US copyright litigation history. According to a press release provided to Ars Technica, each author will receive an estimated $3,000 per work that Anthropic stole, pending court approval. However, if a large number of claims are submitted, the final figure could be higher. "We believe this settlement is a significant step forward in protecting authors' rights and holding AI companies accountable for their actions," said Justin Nelson, a lawyer representing three authors who initially sued to spark the class-action lawsuit. "This case highlights the importance of ensuring that AI models are trained on data that has been obtained lawfully." The settlement follows a years-long investigation into Anthropic's use of copyrighted materials to train its AI models. The company had claimed that it was using public domain works, but an investigation by the authors revealed that many of the works were in fact copyrighted. "This case is a wake-up call for the tech industry," said Andrea Bartz, one of the authors involved in the lawsuit. "We need to ensure that AI companies are not profiting from stolen intellectual property." The settlement is seen as a major victory for authors and creators who have long been concerned about the impact of AI on their livelihoods. "This case shows that we will no longer tolerate the exploitation of our work by tech companies," said Kirk Wallace Johnson, another author involved in the lawsuit. "We will continue to fight for our rights and ensure that our work is respected." The settlement still requires court approval, which may be granted as early as this week. However, the ultimate decision may be delayed until 2026. Background: Anthropic's use of copyrighted materials to train its AI models has been a contentious issue in the tech industry. Many experts have raised concerns about the ethics of using stolen intellectual property to develop AI models. "This case highlights the need for greater transparency and accountability in the development of AI," said Charles Graeber, an author involved in the lawsuit. "We need to ensure that AI companies are not profiting from stolen work." Implications: The settlement has significant implications for the tech industry and the future of AI development. "This case sets a precedent for the use of copyrighted materials in AI training," said Nelson. "It shows that authors will no longer tolerate the exploitation of their work by tech companies." The settlement also raises questions about the role of AI in society and the need for greater regulation. "As we continue to develop more advanced AI models, we need to ensure that they are developed with transparency and accountability," said Bartz. "This case is a step forward in ensuring that authors' rights are respected." Next Developments: The settlement still requires court approval, which may be granted as early as this week. However, the ultimate decision may be delayed until 2026. In the meantime, experts say that the settlement will have far-reaching implications for the tech industry and the future of AI development. "This case is a wake-up call for the tech industry," said Johnson. "We need to ensure that AI companies are not profiting from stolen intellectual property." *Reporting by Arstechnica.*