Sony Music Publishing and Warner Chappell Music have accused Anthropic of extensive copyright infringement in training its AI models, Claude. The publishers are seeking up to $150,000 for each unlawfully used musical work, according to Music Business Worldwide.
Music Publishers vs. Claude
The lawsuit was filed in the U.S. District Court for the Northern District of California. In addition to Anthropic, the defendants include its CEO Dario Amodei and co-founder Benjamin Mann. Sony and Warner allege that the company systematically uploaded and used copyrighted works to train Claude.
According to the plaintiffs, the case involves tens of thousands of musical compositions, including well-known songs like Ain’t No Mountain High Enough, Livin’ on a Prayer, and Hallelujah.
The publishers also claim that Anthropic used pirated copies of books and other materials, and extracted song lyrics from licensed sites such as Musixmatch and LyricFind.
Hundreds of Thousands of Dollars Per Song
Sony and Warner are demanding a jury trial and compensation of up to $150,000 for each infringed work. An additional $25,000 could be sought for each instance of copyright information removal or alteration.
Theoretically, if infringement is proven for a large number of works, the claims could amount to several billion dollars.
The publishers describe Anthropic’s actions as “one of the largest and most blatant ongoing thefts of intellectual property in history.”
“We disagree with the publishers’ claims and intend to vigorously defend ourselves in court,” a representative for Anthropic told TechCrunch.
Anthropic Previously Faced Similar Allegations
The new lawsuit is another episode in the ongoing conflict between Anthropic and rights holders.
In January 2026, Universal Music Group, Concord, and ABKCO filed a separate $3 billion lawsuit against the company, accusing Claude of infringing copyrights on more than 20,000 musical works.
Earlier, Anthropic faced claims from authors and publishers over the use of pirated book copies for model training. In 2025, the company agreed to pay $1.5 billion to settle that case.
Previously, journalists at 404 Media discovered that Amazon scanned and destroyed rare books for AI training, thus avoiding copyright infringement accusations.
