Stakeholders said Anthropic trained its models on 250-plus Elvis songs and demanded $41 million in compensation.
A new lawsuit over a modern issue, AI copyright infringement, reaches back into the past. The owners of Elvis Presley’s music just sued Anthropic for $41 million, claiming the AI company trained its models on 274 Elvis songs without permission, The New York Times reported. That amount for a single artist who died nearly 50 years ago shows how expensive alleged musical plagiarism can be for AI companies.
Stakeholders in Elvis Presley’s back catalog include Authentic Brands Group, Raleigh Music Publishing, multiple trusts and at least one individual. They accuse Anthropic of bypassing DRM protections and removing or altering copyright management information from 247 songs. Those include hits like Don’t Be Cruel, Can’t Help Falling in Love and A Little Less Conversation, according to the lawsuit filed on October 23rd.
“The Elvis Plaintiffs bring this action to enforce these exclusive rights against the Defendant that has built its entire business model based on copyright infringement at a scale never seen before,” lead attorney Richard Busch wrote. “The Elvis Plaintiffs’ iconic collective catalog contains some of the most influential and enduring musical compositions in human history.” The complaint notes that Elvis sold over a billion records and that his works “are some of the most valuable in the world.”
Like OpenAI and other AI companies, Anthropic is facing multiple lawsuits for training their models on copyrighted music and other materials. Sony and Warner recently sued Anthropic for “blatant violation” of copyright law involving thousands of works. Prior to that, Universal Music and Concord Music Group filed a similar claim against Anthropic for illegally downloading and training on 20,000 copyrighted songs. Each lawsuit is claiming damages in the billions of dollars.
The Elvis suit, though, is notable because it centers on a single musician who died decades before AI was a thing (though some have tried to recreate him as a robot). Anthropic has previously claimed that training its AI models on copyrighted music constitutes “fair use.” The company told the NYT in a brief statement that it disagrees with the claims and intends to defend itself in the case.
