Anthropic is actively opposing major music publishers' preliminary injunction request regarding their AI chatbot Claude's alleged copyright infringement of protected musical works.

The publishers seek two main injunctions:

• Remove protected works from Claude's training data
• Block protected lyrics from appearing in Claude's outputs

Key points from Anthropic's opposition:

1. Fair Use Defense

• Using copyrighted works to train LLMs constitutes fair use
• Training data usage is "transformative" under fair use doctrine
• Any potential damages could be compensated monetarily

2. Technical Context

• Claude learns from "trillions of tiny textual data points"
• Training data likely includes some copyrighted works
• Research details predated Claude's commercial release by nearly a year

3. Protective Measures

• Implemented "broad array of safeguards" to prevent copyright violations
• No reasonable expectation of continued infringement
• Disputes claims of ongoing market and licensing harm

Supporting the opposition, Anthropic co-founder Jared Kaplan provided detailed testimony about Claude's training process and technical specifications.

The case (5:24-cv-03811) remains ongoing, with reports suggesting a significant portion may be dismissed in the near future. The outcome could set important precedents for AI training data usage and copyright law.

Related Articles

Previous Articles