The Pulse
Reservoir Seeks Up to $338 Million in Suno and Anthropic Suits
Reservoir Media filed separate copyright lawsuits against Suno and Anthropic on October 21, 2026, alleging the companies used its music without permission to train AI models. The complaints seek damages that could total about $337.8 million
AI.info Team · Written with AI · fact-checked ·
Reservoir Media’s two lawsuits put about $337.8 million in potential statutory damages at issue, based on the maximum amounts the company requests across its claims against Suno and Anthropic. Reservoir filed the complaints on October 21, 2026, in the U.S. District Court for the Northern District of California. The sums are not awards or established losses: the complaints seek actual damages and defendants’ profits, or statutory damages as an alternative, alongside separate claims under the Digital Millennium Copyright Act.
Two complaints, two kinds of AI training
Reservoir Media Management, its UK subsidiary and Shapiro, Bernstein & Co. filed one case against Suno and another against Anthropic. The Suno complaint lists 1,050 musical compositions whose U.S. copyrights Reservoir says it owns outright or exclusively controls. Its examples include Joni Mitchell’s “Both Sides Now,” Johnny Cash’s “Ring of Fire,” Madonna’s “Papa Don’t Preach” and Evanescence’s “Bring Me to Life.”
The separate Anthropic case names 853 compositions and focuses on their lyrics, which Reservoir alleges were copied to train the Claude models. It also accuses Claude of generating copies or derivative versions of lyrics in response to prompts. The Anthropic complaint names works including “Big Yellow Taxi,” “Lush Life” and “Papa Don’t Preach.”
“when we believe their work has been exploited without authorization, we will hold AI platforms legally accountable to ensure those rights are upheld.”
Golnar Khosrowshahi, Reservoir founder and CEO
Reservoir’s case against Suno points to scraped music
Reservoir alleges that Suno used music copied from streaming and lyrics platforms to train its models, without a licence. The complaint cites alleged Suno source-code records and Bright Data’s scraping services, and says a training-data file called “youtube_music” contained more than two million music clips, totaling about 113,879 hours. Those details appear as allegations in Reservoir’s filing, not findings by a court.
The publisher also argues that Suno’s licensing agreements with Warner Music Group, BMG and Believe show that a market exists for licensed music used in training. Its complaint says those deals “all but admitted that there is a functioning market for licensing music and sound recordings for training data.” Reservoir says Suno did not obtain permission to use its catalog.
The $337.8 million figure is a ceiling calculation
For Suno, Reservoir seeks $157.5 million in statutory copyright damages as an alternative to actual damages and profits, calculated at up to $150,000 for each of the 1,050 works. The complaint also lists $2.625 million for alleged circumvention of access controls and $26.25 million for alleged removal of copyright-management information. Together, those figures total $186.375 million.
For Anthropic, the complaint seeks up to $127.95 million in alternative statutory copyright damages for 853 works, plus $2.1325 million and $21.325 million on the two DMCA claims. That adds up to $151.4075 million; combined with the Suno figures, the pleaded amounts total about $337.8 million. Reservoir also asks for an injunction and for unauthorized copies to be impounded or destroyed, including copies it says remain in training datasets, internal repositories, model weights and server infrastructure.
Copyright claims remain allegations
The complaints argue that copying works for training and producing outputs based on them infringed Reservoir’s rights, and that fair use does not excuse the alleged conduct. The filing against Suno links that argument to the company’s licensing agreements; the Anthropic complaint centers on lyrics, Claude’s training and alleged lyric outputs. Neither complaint resolves whether the training uses or outputs violate copyright law.
For songwriters and publishers, the cases test whether using catalog material to build commercial AI systems requires permission and payment, and what remedies apply if a court finds infringement. The court will have to assess the companies’ conduct, the evidence about the works and training processes, and Reservoir’s claims under copyright and the DMCA. At filing, those questions remained allegations for the court to decide.