Anthropic sued over using Elvis Presley music to train Claude AI

The Elvis Presley lawsuit

Anthropic, the company responsible for the Claude AI models, is facing a new legal challenge regarding its data collection practices. A unit of Authentic Brands Group, which manages the licensing for Elvis Presley's music, filed a lawsuit on October 23. The complaint alleges that the tech firm used the music catalog without permission.

The lawsuit specifically names 274 copyrighted musical compositions belonging to the legendary artist. These include well-known tracks such as "Can't Help Falling in Love," "Don't Be Cruel," and "A Little Less Conversation." The legal action was co-filed by other entities that hold rights to Presley's work.

According to a report by Gadgets 360, the plaintiffs are aiming for significant financial penalties. They are seeking $150,000 for every single one of the 274 tracks cited in the complaint. This represents a major financial risk for the AI company as it attempts to scale its technology.

Allegations of mass infringement

The legal team representing the Elvis estate is making a strong case against the scale of AI data scraping. Richard Busch, the lead attorney for the lawsuit, stated that the plaintiffs want to enforce exclusive rights against a business model built on "copyright infringement at a scale never seen before." This quote reflects the core of the dispute over how much data is being ingested by these systems.

The list of works cited in the complaint is extensive and covers many of the artist's most famous hits. In addition to the primary tracks mentioned, the lawsuit includes songs like "All Shook Up," "Return to Sender," and "(You're The) Devil in Disguise." The plaintiffs argue that these works were used to build the foundation of the AI's capabilities.

Anthropic has not yet released a public statement regarding these specific allegations. However, the company is already navigating a complex landscape of intellectual property disputes. This case adds another layer of scrutiny to how they source the information used to teach their models.

A pattern of AI copyright battles

This lawsuit is not an isolated incident in the tech industry. It is part of a broader trend where various creative industries are pushing back against AI companies. For example, Universal Music Group has already initiated legal action against AI firms for using copyrighted music in their training sets.

The dispute over music is one of many fronts where AI companies are facing pushback from content creators. Other industries, including the publishing world, have seen similar legal challenges. These cases often center on whether using copyrighted material to train a model constitutes fair use or a violation of ownership.

Anthropic has already dealt with significant legal hurdles in other areas of creative production. Earlier this year, a federal judge in San Francisco approved a $1.5 billion settlement for a class action lawsuit. That suit involved a group of authors who claimed their books were misused to train the Claude models.

The mechanics of model training

To understand why this lawsuit matters, it is helpful to look at how large language models are built. These models require massive amounts of data to learn patterns in human language, music, and art. During the training phase, the AI processes these files to improve its ability to generate content.

The core of the legal argument rests on whether this processing constitutes a new creation or a theft of the original work. The plaintiffs argue that by using the music to build a commercial product, Anthropic is infringing on the exclusive rights of the owners. They believe the technology is profiting directly from the creative output of the artist.

If the court finds in favor of the music licensing companies, it could change how AI companies source their training data. Developers might be forced to pay licensing fees for every piece of media they use. This could make it more expensive and legally complicated to develop high-performing AI models.

What comes next for AI law

The outcome of this lawsuit could have a ripple effect across the entire entertainment and tech sectors. If the plaintiffs succeed in their claim for $150,000 per track, it could set a high price floor for music licensing in the AI era. Other artists and labels may follow suit to secure similar payouts for their catalogs.

For Anthropic, the case represents a significant hurdle in its quest to provide a safe and useful AI tool. They must balance the need for diverse training data with the legal requirements of copyright law. The company will need to determine how to continue improving Claude while avoiding further litigation from rights holders.

The tech community will be watching this case closely to see how the courts define the boundaries of AI training. As more lawsuits are filed, the legal framework for artificial intelligence will continue to evolve. This case serves as a reminder that the intersection of new technology and old property rights is a primary battleground for the next decade.