AI music copyright rules: who owns what and how the law is evolving

01The Current State of AI Music Copyright Law

In the 2020s, the rapid advancement of deep learning-based generative artificial intelligence models raised questions about the copyright status of AI-generated works and whether copyright infringement occurs during model training. A copyright is a type of intellectual property giving its owner the exclusive legal right to copy, distribute, adapt, display, and perform a creative work. The intersection of these established rights with AI-generated music remains an evolving legal landscape with no settled consensus.

02Training Data and Fair Use Questions

Fair use is a doctrine in United States law that permits limited use of copyrighted material without first acquiring permission from the copyright holder. AI music models are trained on vast datasets that often include copyrighted songs, raising the question of whether this constitutes fair use or infringement. Courts are weighing whether transformative use applies when the training data is not directly reproduced but rather used to learn statistical patterns of musical composition.

03Who Owns AI-Generated Compositions

Under current US copyright law, works generated entirely by AI without human authorship may not qualify for copyright protection, as copyright requires human authorship. This creates a legal gray area: if a human prompts an AI to generate music and then selects, arranges, and modifies the output, does that human become the author? The Copyright Office has issued guidance suggesting sufficient human creative contribution is required, but the boundary remains unclear.

04Voice Cloning and Right of Publicity

Voice cloning technology allows AI to recreate a singer's voice with striking accuracy, raising distinct legal questions beyond copyright. The right of publicity protects individuals from unauthorized commercial use of their identity, including their voice. Several high-profile cases involving AI-generated songs mimicking famous artists have tested this legal framework, leading to demands for stronger protections against non-consensual voice replication.

05How Record Labels Are Responding

Major record labels have filed lawsuits against AI music platforms alleging unauthorized use of copyrighted recordings for training. Some labels are simultaneously exploring licensing agreements that would allow controlled use of their catalogs. The industry is also pushing for legislative solutions that would establish clear rules for AI training data usage and strengthen protections for artists against voice cloning.

06Landmark Cases and Settlements

Several cases are shaping the AI music copyright landscape. Lawsuits against AI music generators allege direct infringement through training on copyrighted works. Settlements and licensing deals have begun to establish practical norms, though judicial precedent remains sparse. The outcomes of pending litigation will likely determine whether AI music platforms must license training data or can claim fair use protections.

07What Creators Should Know to Protect Their Work

Creators navigating the AI music landscape should register their works with the Copyright Office, understand that AI-generated components may not be protectable, and consider contractual provisions addressing AI usage. Artists concerned about voice cloning should be aware of right of publicity protections in their jurisdiction. Staying informed about evolving case law and legislative proposals is essential, as the legal framework is likely to shift significantly in the coming years.