Several major music publishers, including Sony Music and Warner Chappell, filed a lawsuit against artificial intelligence company Anthropic on August 28, 2026, accusing the company of conducting what they describe as 'one of the largest and most blatant ongoing thefts of intellectual property in history'. According to the latest reports, the lawsuit was filed in the US District Court for the Northern District of California and names co-founders Dario Amodei and Benjamin Mann as co-defendants alongside the company. The companies allege that Anthropic conducted a 'brazen campaign of illegally torrenting, scraping, and downloading copyrighted works on a massive scale' to develop and operate their 'Claude' series of AI models. As the lawsuit notes, another district court recently described Anthropic's actions as 'straightforward piracy but at massive scale,' which the music publishers say is perhaps the most accurate description of the AI industry imaginable. Anthropic has responded to the allegations, with a company spokesperson stating 'We disagree with the publishers' claims and we intend to defend ourselves robustly in court,' as reported by Business Insider.
The lawsuit claims Anthropic collected 'tens of thousands' of copyrighted songs, including popular works such as 'Eye of the Tiger,' 'September,' 'Uptown Funk,' 'Hallelujah,' 'Taylor Swift's 'Paper Rings,' and Mariah Carey's 'All I Want for Christmas Is You.' The companies alleged that Anthropic pirated their works using two digital archives, Library Genesis and Pirate Library Mirror, and also scraped lyrics from websites including Musixmatch and LyricFind. According to the complaint, Mann used the BitTorrent protocol in June 2021 to download at least five million pirated books from Library Genesis, known as LibGen, and Anthropic employees torrented at least two million more from a successor site called Pirate Library Mirror in July 2022. The music publishers claim that Anthropic 'harvested innumerable copyrighted works' in 'blatant violation of copyright law' and then 'systematically copied' them multiple times to train its models, 'inflicting immense harm on Music Publishers and the songwriters they proudly represent.' The publishers further argue that Claude can reproduce copyrighted lyrics either verbatim or substantially as written, and they argue that such outputs could compete with music created by human songwriters.
The publishers are seeking a jury trial and are reportedly seeking statutory damages from the AI firm, including up to $150,000 for each composition it used to train the chatbot. The complaint brings four specific claims against the defendants. The first alleges direct copyright infringement through torrenting against all three defendants, while the second charges Amodei and Mann personally with contributory infringement for directing and approving the torrenting activity. The third and fourth counts, brought against Anthropic alone, allege direct infringement through scraping, downloading, destructive scanning, model training, and AI outputs, plus removal or alteration of copyright management information. The publishers also seek $25,000 for each instance in which Anthropic removed copyright management information from each work. According to the complaint, the publishers are seeking 'an order requiring Anthropic to destroy all infringing copies of their works under court supervision' and 'an accounting of Anthropic's training data, training methods, and the specific lyrics and other copyrighted works used to train its models.' The potential financial exposure could run into billions of dollars, although any final damages would depend on the court's findings.
This lawsuit represents the latest in a series of copyright infringement cases against Anthropic in the music industry. Earlier this year, Concord Music Group and Universal Music Group separately sued Anthropic, accusing it of illegally downloading more than 20,000 copyrighted songs for AI training purposes and seeking more than $3 billion in damages. The company has also faced copyright lawsuits from groups of authors who claimed Anthropic used pirated copies of their protected works to train its AI models. Anthropic previously settled a record-setting $1.5 billion lawsuit with authors over similar copyright violations, highlighting the significant financial stakes in these cases. The music publishers allege that previous settlements have not been sufficient to deter such conduct, with the current lawsuit representing a 'brazen campaign of illegal torrenting' that exceeds the scale of previous violations. This latest lawsuit builds on previous cases, including the Bartz v. Anthropic case where a group of authors accused the company of using copyrighted works to train products like Claude, with Anthropic ultimately ordered to pay $1.5 billion after a judge ruled that while it was legal for the AI lab to use copyrighted works, it was not legal to acquire that content through piracy. The publishers specifically reference Anthropic's $1.5 billion settlement with book authors in September 2025 over similar torrenting conduct, stating that 'Anthropic clearly considers that to be just the cost of doing business given that its entire business model continues to be built on copyright theft.'
Anthropic is not alone in facing legal challenges over AI training data across the industry. OpenAI has been sued by WikiHow in August 2026, alleging that more than 11,000 instructional articles were scraped without permission to train ChatGPT. Meta has faced lawsuits from publishers including Elsevier, Cengage, Hachette, Macmillan, and McGraw Hill in May 2026, alleging that copyrighted books and journal articles were used to train its Llama models. Google has also been targeted by publishers and authors in a class-action lawsuit accusing the company of using copyrighted works to train its Gemini AI platform. These cases collectively represent a 'brazen campaign of illegal torrenting' that exceeds the scale of previous violations, with music publishers arguing that AI companies must obtain copyrighted creative works lawfully and compensate rights holders where required.
