Key Takeaways
- New copyright lawsuit: Elvis Presley music rights holders sued Anthropic in a California federal court on October 23, 2026.
- 274 songs involved: The complaint alleges that Anthropic used copyrighted musical compositions associated with Presley to train its Claude AI models without authorization.
- Damages sought: The plaintiffs are seeking approximately $41.1 million for alleged infringement of the 274 works, with additional claims reportedly bringing the total requested damages to around $48.6 million.
- Alleged AI-generated copies: The complaint claims Claude generated song rewrites that retained recognizable structures and hooks despite changes to the wording.
- Anthropic’s response: The company has disputed the allegations and said it intends to defend itself.
- Wider implications: The case adds to the legal debate over whether and how AI companies can use copyrighted creative works to train their models.
What Is the Elvis Presley Copyright Lawsuit Against Anthropic?
The lawsuit was filed by entities that own or control rights to Elvis Presley’s musical compositions. The plaintiffs include ABG EPE IP LLC, associated with Authentic Brands Group, alongside other rights holders.
According to Reuters, the case, ABG EPE IP LLC v. Anthropic PBC, was filed in the U.S. District Court for the Northern District of California under case number 5:26-cv-12473.
The plaintiffs allege that Anthropic used lyrics from 274 copyrighted musical compositions without obtaining permission or paying the relevant rights holders. The works named in the complaint include well-known songs such as “Can’t Help Falling in Love,” “All Shook Up,” “Don’t Be Cruel” and “Return to Sender.”
The lawsuit also alleges that Claude can generate unauthorized derivative versions of copyrighted songs. The plaintiffs argue that the issue extends beyond the material used to train the model to the content the AI system can produce in response to user requests.
These remain allegations in a newly filed case. The court has not determined whether Anthropic infringed the plaintiffs’ copyrights.
How Much Money Are Elvis Presley Rights Holders Seeking?
The plaintiffs are seeking substantial financial damages, arguing that the alleged use of their copyrighted works violated their exclusive rights.
| Legal claim or measure | Reported amount |
|---|---|
| Number of musical compositions at issue | 274 |
| Requested damages per work | Up to $150,000 |
| Calculated amount for 274 works | $41.1 million |
| Reported additional claims | Approximately $7.535 million |
| Total reported damages request | Approximately $48.635 million |
Sources: Bloomberg Law and Los Angeles Times.
The figures represent the plaintiffs’ requested relief, not a court award. The ultimate amount, if any, would depend on the legal claims, evidence, applicable law and court proceedings.
The case also includes allegations involving the circumvention of digital access controls and the removal of copyright information. Those claims are separate from the basic allegation that copyrighted lyrics were used for AI training.
Why the Lawsuit Targets Claude’s Training Data
Large language models learn statistical patterns from training material. For a model such as Claude, that process can involve large collections of text, code and other data used to develop its ability to understand and generate language.
The plaintiffs contend that Anthropic used copyrighted Elvis Presley compositions without permission as part of this process. They argue that the company’s use of the works violated copyright holders’ exclusive rights.
The central legal question is not simply whether an AI system has encountered copyrighted material. It is whether the particular collection, copying, use and resulting outputs violate applicable copyright law.
AI companies have argued in other cases that some uses of copyrighted works for model training may qualify as fair use under U.S. law. That defense is fact-specific, and a decision in one case does not automatically settle every dispute involving different works, datasets or uses.
The Presley lawsuit therefore adds another set of facts to a rapidly developing legal debate. Its outcome could help clarify how courts assess the rights of music publishers and other copyright owners when their works are used in training generative AI systems.
The Complaint Also Challenges Claude’s Generated Outputs
The plaintiffs allege that Claude can produce versions of copyrighted songs that remain too close to the originals.
According to reporting by the Los Angeles Times, the complaint describes outputs that allegedly retained recognizable song structures, hooks and arrangements even when some wording was changed.
The plaintiffs also argue that Anthropic’s safeguards against copyright-infringing outputs were ineffective. They contend that the company continued to face problems after introducing measures intended to prevent the reproduction of protected material.
This allegation raises a separate issue from the use of songs in training. A model’s training process and its responses to individual prompts are not legally identical questions. Evidence that a model produces a similar song does not, by itself, prove precisely which training material it used or whether the training process was unlawful.
Nevertheless, the allegations highlight the importance of testing AI systems for outputs that reproduce protected creative expression. Companies developing generative models must consider not only their training-data practices but also how their products respond to requests involving copyrighted material.
How the Case Fits Into Anthropic’s Wider Copyright Disputes
Anthropic has faced copyright litigation from authors and music publishers over the use of protected material in AI development.
In 2025, the company agreed to a $1.5 billion settlement in a class-action lawsuit brought by authors who alleged that Anthropic used pirated copies of their books to train its AI systems. That settlement was a separate legal matter and does not automatically determine the outcome of the Presley case.
Music publishers have also pursued claims against AI companies over training data and generated content. The Presley lawsuit is notable because it focuses on a specific artist’s catalogue and involves rights holders seeking to enforce control over hundreds of musical compositions.
The distinction matters because music rights can be divided among different parties. Rights in a musical composition—such as its lyrics and melody—are separate from rights in a particular sound recording. The current case concerns alleged infringement involving musical compositions, rather than a blanket finding that every recording of Presley has been used unlawfully.
The expanding number of disputes suggests that licensing arrangements for AI training could become an increasingly important part of the music industry’s relationship with technology companies.
What the Lawsuit Could Mean for AI Companies and the Music Industry
The case could influence how AI developers assess training datasets, document permissions and respond to requests for material resembling existing songs.
If courts find that particular uses of copyrighted compositions require permission, AI companies may face greater pressure to negotiate licenses with music publishers and rights holders. Such arrangements could create new revenue opportunities for the music industry while giving AI developers a clearer framework for obtaining training material.
On the other hand, a ruling favorable to Anthropic on particular claims could help define circumstances in which AI training or certain outputs are legally permissible. The result would depend on the court’s reasoning and the scope of any decision.
For musicians, publishers and estates, the case highlights the challenge of protecting creative works when AI systems can generate new material based on patterns learned from large datasets. For developers, it reinforces the need to consider copyright risks throughout the development and deployment process.
The Bigger Picture
The dispute over Elvis Presley’s music is part of a wider question about how generative AI should interact with copyrighted culture. AI models can help people create, analyze and transform content, but those capabilities also raise concerns about compensation, permission and the potential reproduction of existing works.
There is no single legal rule that resolves every training-data dispute. Courts must consider the specific works, conduct, defenses and claims involved. The Presley case could add important evidence to that debate, particularly around music catalogues and allegations that AI systems generate close imitations of copyrighted songs.
FAQs
1. Why are Elvis Presley rights holders suing Anthropic?
The plaintiffs allege that Anthropic used lyrics from 274 Elvis Presley musical compositions to train Claude AI models without permission. They also allege that Claude can produce unauthorized rewrites resembling protected songs.
2. How much compensation are the plaintiffs seeking?
The plaintiffs reportedly seek up to $150,000 for each of the 274 works, amounting to $41.1 million, alongside additional claims. Reports put the combined request at approximately $48.6 million. These are requested damages, not an award.
3. Has Anthropic admitted to copyright infringement?
No. Anthropic has disputed the allegations and said it intends to defend itself. The lawsuit remains a legal dispute, and the claims have not been established by a court.
4. Does using copyrighted songs to train AI automatically violate copyright law?
Not automatically in every circumstance. The legal analysis depends on the specific conduct, the copyrighted material, the applicable law and any defenses, including fair use where relevant. The Presley lawsuit has not yet resolved those questions.
5. Could this lawsuit change how AI companies train their models?
Potentially. The case may influence future discussions about licensing, training-data documentation and safeguards against outputs that reproduce protected material. Its broader impact will depend on the court’s eventual decisions and whether the parties reach a settlement.
Looking Ahead
The immediate next steps will be determined by the court proceedings, including Anthropic’s response to the complaint and how the plaintiffs substantiate their claims. The case could take time to resolve, and any eventual settlement or ruling would need to be assessed in light of the specific legal issues involved.
For the AI industry, the dispute is another reminder that access to large datasets is not only a technical challenge but also a legal and commercial one. As generative AI expands into music, writing and other creative fields, licensing arrangements, transparent data practices and effective output safeguards are likely to remain central to negotiations between technology companies and copyright owners.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
