Anthropic Hit With $3B Lawsuit Over Alleged AI Music Piracy

Universal Music Group, Concord and ABKCO escalate their legal battle against Anthropic, accusing the AI company of mass-scale copyright infringement involving over 20,000 songs.

$3B Copyright Bombshell Dropped on Anthropic

A heavyweight coalition of music publishers has launched what may become the most consequential copyright lawsuit the AI industry has faced. Universal Music Group, Concord Music Group and ABKCO Music filed a $3 billion complaint against Anthropic on Wednesday (January 28), alleging "persistent and brazen" copyright infringement of over 20,000 songs.

The Core Allegations

1. Mass-Scale Piracy Operation

The lawsuit claims Anthropic co-founder Benjamin Mann personally used BitTorrent to download millions of pirated books - including songbooks containing copyrighted compositions - from platforms like LibGen. Court documents allege this content was then fed into Anthropic's AI training pipeline.

2. Direct Targeting of Founders

In a rare aggressive move, the complaint names Anthropic co-founders Dario Amodei and Benjamin Mann as individual defendants. It alleges Amodei "personally authorized and discussed" the torrenting activity.

3. Ongoing Infringement in AI Outputs

Publishers claim Anthropic continues to violate copyrights by using lyrics in: - Training newer Claude AI models - User-generated outputs

Why This Case Matters

This represents a dramatic escalation from publishers' 2023 lawsuit against Anthropic, which covered just 500 works with $75 million in potential damages. Three key factors make this case pivotal:

1. Scale: 20,000+ songs at stake 2. Direct Evidence: Alleged torrenting activity documented 3. Precedent: Could set standards for AI training data sourcing

The Smoking Gun

Publishers say they only uncovered the alleged torrenting operation after a separate case (Bartz v. Anthropic) exposed Anthropic's reliance on pirate libraries. A July 2025 ruling by Judge William Alsup publicly revealed details of Anthropic's data sources.

What's Next

The Northern District of California will now consider whether to: - Allow the case to proceed - Determine if statutory damages could reach $3 billion - Potentially set new boundaries for AI training data practices

This case could become the largest non-class action copyright lawsuit in U.S. history - and a watershed moment for AI and creative rights.