Preliminary Victory for Anthropic: AI Challenges Music Industry Giants

Tech
By 24matins.uk,  published 26 March 2025 at 17h58, updated on 26 March 2025 at 17h58.
Tech

A California judge has dismissed music publishers' lawsuit against Anthropic, reigniting a heated debate on the fair use of copyrighted content for training artificial intelligence.

In a burgeoning case concerning copyrights in artificial intelligence, AI firm Anthropic has persuaded a California federal judge to dismiss a preliminary injunction that would have prevented the use of song lyrics owned by Universal Music Group and other music publishers to train its chatbot, Claude.

Music Publishers Strike Back

In 2023, UMG, Concord, and ABKCO sued Anthropic, alleging the company had “infringed their copyrights by using lyrics from at least 500 songs by artists like Beyoncé, the Rolling Stones, and the Beach Boys” to train its chatbot to respond to human commands.

This trial is among several asserting that copyrighted works from authors, news media, visual artists, and others have been used without consent or compensation to develop AI-based products.

Technology firms, including OpenAI, Microsoft, and Meta Platforms, maintain that their systems use copyrighted material “fairly” under U.S. copyright law by studying it to learn how to create transformative new content.

Judge’s Ruling

Judge Eumi Lee dismissed the publishers’ claim that Anthropic’s use of their lyrics caused irreparable harm by diminishing their licensing market. “The publishers are essentially asking the Court to outline the boundaries of a licensing market for AI training where the fair use issue remains unresolved,” she stated.