Elvis Presley estate sues Anthropic over unauthorized music training

The lawsuit against Anthropic

Authentic Brands Group, the organization that owns the rights to Elvis Presley’s music and likeness, has filed a lawsuit against the AI company Anthropic. The legal action, filed on Friday, accuses the developer of using 274 of Presley's songs to train its models without obtaining permission. This move marks a significant shift in the legal landscape for artificial intelligence.

While other major music labels and the Recording Industry Association of America (RIAA) have pursued copyright claims against AI firms, this specific case is notable for its origin. It represents one of the first times an artist's estate has filed a direct legal challenge against an AI developer. The lawsuit focuses on the unauthorized use of a massive, iconic music catalog to build commercial AI products.

The plaintiffs are seeking $150,000 for each of the 274 infringed tracks, which totals approximately $41 million. Lead attorney Richard Busch stated in the filing that the plaintiffs are moving to enforce exclusive rights against a business model built on copyright infringement at an unprecedented scale. The lawsuit argues that the AI model, Claude, has produced infringing content for virtually every song available on the internet.

Specific works and allegations

The lawsuit highlights several of Presley's most famous works as being included in the unauthorized training data. These include well-known tracks such as "Can’t Help Falling in Love," "All Shook Up," and "A Little Less Conversation." The plaintiffs argue that these specific works were ingested into the system to help the AI learn the nuances of Presley's music.

The legal filing also takes aim at the safety measures Anthropic has put in place to manage its outputs. The lawsuit claims that even the institution of guardrails—which are tools meant to limit what an AI can generate—have been ineffective. The plaintiffs argue that Anthropic knew these guardrails failed to stop the creation of infringing music and continued to engage in the practice regardless.

According to the complaint, Anthropic knowingly and willfully engaged in copyright infringement of musical works. The lawsuit suggests that the company's actions demonstrate a pattern of behavior rather than an accidental oversight. By using these songs to train its models, the company is accused of leveraging the value of Presley's legacy to power its own technology.

How AI music training works

To understand the core of this dispute, it is necessary to look at how AI music platforms operate. These systems work by analyzing massive amounts of recorded music to learn various elements of the craft. This includes identifying the specific sounds, structures, and expressions that make a song unique to an artist or genre.

Once the AI has processed this data, it can allow users to create their own AI-generated music through simple text or speech prompts. By learning from existing human-made music, the AI can mimic styles that were not explicitly programmed by humans. This process allows the technology to generate new compositions that sound similar to the original training data.

Musicians have expressed deep concerns about this process because it allows their work to be used without their consent. There is a growing fear that AI-generated "slop"—a term for low-quality, mass-produced content—could eventually replace human-made songs. Some bots have even reached the point of impersonating living artists on various music streaming platforms.

Impact on the music industry

The music industry is currently grappling with the rapid rise of AI-generated content and its impact on creators. Many musicians fear that their voices and unique styles are being harvested to create products that compete with their own careers. This has led to a push for stricter regulations and clearer legal boundaries regarding how training data is sourced.

The scale of the problem is becoming visible to major platforms as well. For example, the streaming giant Spotify recently announced it had removed about 50 million songs considered AI slop from its recommendations over the past year. This indicates a growing effort to manage the influx of automated content on popular streaming services.

The outcome of the Presley estate's lawsuit could set a broad precedent for how AI companies are allowed to use existing music catalogs. If the court rules in favor of the estate, it could force AI developers to secure licenses for every piece of music they use in training. This would fundamentally change the cost and methodology of developing large-scale generative models.

What comes next for AI law

This case will likely be watched closely by both the technology sector and the entertainment industry. Because it involves a high-profile artist's estate, the legal arguments regarding copyright will be scrutinized by courts across the country. The final ruling may dictate how much of the internet's creative output can be used as free fuel for new technologies.

If the court finds that the use of these songs constitutes willful infringement, it could lead to significant financial penalties for AI developers. It may also force these companies to develop more transparent methods for sourcing their training data. This could involve creating licensing pools where artists are paid for the use of their work in AI models.

The litigation serves as a warning to other rights holders about the risks of AI training. As more artists and estates consider their legal options, the pressure on AI companies to establish clear, fair, and legal ways to use copyrighted material will only increase. The case highlights the tension between technological progress and the protection of human creativity.