“Plaintiff Music Publishers, a group of the world’s leading music publishers, bring this action to hold accountable the culprits behind one of the largest and most blatant ongoing thefts of intellectual property in history,” starts the allegation that music publisher’s Sony Music Publishing and Warner Chappell Music have made against Anthropic, an American artificial intelligence company known for creating a family of AI assistants in a 48-page complaint. In a significant legal battle, Sony Music and Warner Chappell have sued Anthropic, accusing the AI company of engaging in a “brazen campaign” to steal intellectual property. The lawsuit alleges that Anthropic used copyrighted music and other content to train its AI model, Claude, through unauthorized methods, including torrenting and scraping.
The music publishers are those who manage the copyright of songs for songwriters and composers. Some of the compositions that have been allegedly misused include classics such as Mariah Carey’s All I Want for Christmas is You and Survivor’s Eye of the Tiger. Plaintiffs are seeking $150,000 in damages per work along with $25,000 per copyright violation which puts compensation in the realm of billions of dollars, possibly making this a multibillion-dollar lawsuit.
The lawsuit alleges that Anthropic unlawfully trained its models using “tens of thousands” of copyrighted musical compositions owned by music publishers. From this, Claude models allegedly have created identical or nearly identical copies of the copyrighted work in their response to users. On the other hand, Anthropic takes the position that training Claude on copyrighted content allows for the production of AI-generated lyrics that may ultimately compete with human-made songs and denies the accusations and intends to defend themselves in court.
Notably, this is not the first intellectual property lawsuit that has been filed against Anthropic. Earlier this year, in a similar case, Anthropic reached a $1.5 billion settlement with book publishers where Anthropic was accused of training models on stolen works.
What is copyright infringement?
Generally, copyright infringement refers to the act of unlawful copying of material under intellectual property law and interferes with the right of intellectual property ownership. In this copyright infringement matter, the plaintiffs, here the Music Publishers would need to establish valid copyright ownership and the copying of elements of the original copyrighted work.
Why does this matter?
When artists own copyrightable work and their works are used without their permission, this denies copyright owners the recognition and valuable attribution to their own works. In this lawsuit, Anthropic used thousands of copyrighted works to train Claude, which the publishers describe as blatant theft. For businesses that utilize copyrighted works, it is vital for the businesses to validate data provenance before model training. If small business owners use someone’s protected work, even accidentally, this can lead to expensive legal penalties that may even threaten the company’s existence. For example, adding copyrighted songs to social media posts without permission may be an accidental violation that leads to legal issues. However, using tracks that are specifically cleared for commercial use would be acceptable. On a larger scale, risk teams and AI product teams may benefit from keeping evidence for dataset sources and acquisition methods to ensure there is adequate proof for possible future discrepancies.
The impact of this case is likely to reach far beyond the courtroom. It highlights the growing need for AI developers and content creators to work together to establish clear and fair standards for the use of creative works in AI development. As technology and creative industries continue to intersect, greater collaboration and transparency could help create a framework that protects creators’ rights while allowing AI innovation to continue. Ultimately, finding this balance will be essential to building a sustainable relationship between technological advancement and the creative community.
The lawsuit highlights a broader and increasingly important conflict between the creative industries and the rapid development of artificial intelligence. As AI systems become more capable of analyzing, reproducing, and generating content based on existing works, questions about ownership, consent, and fair use are becoming harder to ignore. The case underscores the need to carefully balance technological innovation with meaningful protections for creators and copyright holders whose work may be used to develop AI systems.
Ultimately, the case raises a fundamental question for the future of artificial intelligence and intellectual property rights: How can we encourage technological innovation while ensuring that intellectual property rights remain protected in an increasingly AI-driven world?
Thanks for reading!