Policy Archive · Episode 09

The Case of the Music Misuse

Explore the real-world policies behind this case. Follow proposed and existing policy, potential gaps, and outcomes as they develop.

• Proposed Policy Proposed federal legislation

  TRAIN Act

  The Transparency and Responsibility for Artificial Intelligence Networks (TRAIN) Act would create a process allowing copyright owners who reasonably believe their work was used to train a generative-AI model to seek records identifying whether their copyrighted material was included in the model’s training data.

  Identifier

  S. 2455 / H.R. 7209

  Jurisdiction

  United States (Federal)

  Status

  Proposed federal legislation

  Dates

  S.2455 was introduced in the U.S. Senate on July 24, 2025 and referred to the Senate Judiciary Committee. H.R.7209, the House companion, was introduced January 22, 2026 and referred to the House Judiciary Committee. The legislation has not been enacted.

  Latest update

  January 22, 2026

  Update date

  2026-01-22

  What it does: The Transparency and Responsibility for Artificial Intelligence Networks (TRAIN) Act would create a process allowing copyright owners who reasonably believe their work was used to train a generative-AI model to seek records identifying whether their copyrighted material was included in the model’s training data.

  Potential impact: For "Music Misuse", the TRAIN Act addresses the first question a musician in this situation could face: How can a creator determine whether an AI system actually trained on their work? The proposal would provide a discovery mechanism. It would not by itself determine whether the use was copyright infringement or establish a general licensing or compensation requirement.

  Sources

  • GovInfo — S.2455
  • Congress.gov — H.R.7209
• Proposed Policy Proposed federal legislation

  Protect Working Musicians Act of 2026

  H.R.8994 would allow qualifying independent music creators to collectively negotiate licensing terms with dominant online music-distribution platforms and companies developing or deploying generative AI.

  Identifier

  H.R. 8994

  Jurisdiction

  United States (Federal)

  Status

  Proposed federal legislation

  Dates

  H.R.8994 was introduced May 21, 2026 and referred to the House Judiciary Committee. It has not been enacted.

  Latest update

  May 21, 2026

  Update date

  2026-05-21

  What it does: H.R.8994 would allow qualifying independent music creators to collectively negotiate licensing terms with dominant online music-distribution platforms and companies developing or deploying generative AI. The bill would also allow qualifying creators to coordinate in refusing to license their music to those companies without violating specified antitrust laws.

  Potential impact: For "Music Misuse", the proposal addresses the relationship between individual musicians and companies seeking access to their music for generative-AI uses. Rather than establishing a government-set licensing price, it would give qualifying independent creators greater ability to negotiate collectively over whether and on what terms their music may be licensed.

  Sources

  • Congress.gov — H.R.8994, Protect Working Musicians Act of 2026
  • GovInfo — H.R.8994 (introduced bill PDF)
• Proposed Policy Proposed federal legislation

  COPIED Act

  Cross-Episode Episode 14

  The Content Origin Protection and Integrity from Edited and Deepfaked Media Act of 2025 (COPIED Act) would establish protections involving digital-content provenance and synthetic-content transparency.

  Identifier

  S. 1396

  Jurisdiction

  United States (Federal)

  Status

  Proposed federal legislation

  Dates

  S.1396 was introduced April 9, 2025 and referred to the Senate Committee on Commerce, Science, and Transportation. It has not been enacted.

  Latest update

  April 9, 2025

  Update date

  2025-04-09

  What it does: The Content Origin Protection and Integrity from Edited and Deepfaked Media Act of 2025 (COPIED Act) would establish protections involving digital-content provenance and synthetic-content transparency. For "Music Misuse", a particularly relevant provision would prohibit certain commercial uses of covered content carrying provenance information for AI training or synthetic-content generation without the copyright owner’s express, informed consent and compliance with applicable terms, including compensation terms established by the copyright owner.

  Potential impact: For musicians and other creators, the proposal could provide greater control over whether protected content is used to train AI systems while using provenance technology to help identify and protect original works. For "Music Misuse", it directly raises the question of whether an AI developer should need a creator’s consent before using protected work for AI training.

  Sources

  • Congress.gov — S.1396, COPIED Act
  • GovInfo — S.1396 (introduced bill PDF)
• Proposed Policy Proposed federal legislation

  CLEAR Act

  The Copyright Labeling and Ethical AI Reporting (CLEAR) Act would require notices to be submitted to the Register of Copyrights concerning copyrighted works used in building generative-AI models.

  Identifier

  S. 3813

  Jurisdiction

  United States (Federal)

  Status

  Proposed federal legislation

  Dates

  S.3813 was introduced February 10, 2026 and referred to the Senate Judiciary Committee. It has not been enacted.

  Latest update

  February 10, 2026

  Update date

  2026-02-10

  What it does: The Copyright Labeling and Ethical AI Reporting (CLEAR) Act would require notices to be submitted to the Register of Copyrights concerning copyrighted works used in building generative-AI models.

  Potential impact: For "Music Misuse", the proposal addresses training-data transparency. Greater disclosure could help copyright owners understand whether protected works have been used in building generative-AI systems. Unlike the COPIED Act’s consent provisions, the CLEAR Act primarily approaches the issue through disclosure and reporting.

  Sources

  • Congress.gov — S.3813, CLEAR Act
  • GovInfo — S.3813 (introduced bill PDF)
• Existing Policy Existing federal law / agency guidance

  U.S. Copyright Law and AI-Generated Works

  Cross-Episode Episode 5

  Existing U.S. copyright law protects original human-authored expression, including musical compositions and sound recordings, and Copyright Office guidance addresses how those principles apply when AI is involved.

  Identifier

  U.S. Copyright Law · Copyright Office AI Reports

  Jurisdiction

  United States (Federal)

  Status

  Existing federal law / agency guidance

  Dates

  Existing U.S. copyright law protects original human-authored expression. The U.S. Copyright Office issued Part 2 (Copyrightability) on January 29, 2025 and a pre-publication Part 3 (Generative AI Training) on May 9, 2025.

  Latest update

  May 9, 2025

  Update date

  2025-05-09

  What it does: Existing U.S. copyright law protects original human-authored expression, including musical compositions and sound recordings. The U.S. Copyright Office has concluded that existing copyright principles can also be applied when AI is involved in creating a work. Using AI as an assistive tool does not prevent copyright protection for human-created expression, but purely AI-generated material—or material for which a human did not exercise sufficient control over the expressive elements—is not itself protected by copyright. Whether a particular human contribution qualifies for protection depends on the circumstances. Existing copyright principles, including infringement and fair use, also provide the legal framework for evaluating uses of copyrighted material in AI training and AI-generated outputs. The Copyright Office’s Part 3 report examines how those principles apply to generative-AI training, including licensing and potential market effects.

  Potential impact: For "Music Misuse", existing copyright law already addresses important parts of the fictional dispute: whether protected musical expression was copied, whether a new use may qualify as fair use, and which human-created elements of a human-AI work may receive copyright protection. The less settled issue is not whether copyright law exists, but how it should apply as generative-AI systems ingest copyrighted works at scale and creators seek greater transparency and control over those uses.

  Sources

  • U.S. Copyright Office — Copyright and Artificial Intelligence
  • U.S. Copyright Office — Part 2: Copyrightability
  • U.S. Copyright Office — Part 3: Generative AI Training (Pre-Publication)
• Policy Gap Partial Gap

  Creator Control Over Copyrighted Works Used for AI Training

  Jurisdiction

  United States (Federal)

  Status

  Partial Gap

  What it does: Existing copyright law can apply when copyrighted works are used in ways that constitute infringement, and emerging federal proposals would provide creators with additional mechanisms involving training-data disclosure, provenance, consent, and collective licensing negotiations. But Judge GPT TV's research did not identify a single, generally applicable federal framework giving copyright owners advance knowledge of, or standardized control over, whether and on what terms their works may be used to train generative-AI systems. Questions involving permission, licensing, compensation, and fair use continue to depend on existing copyright law, developing licensing markets, and the circumstances of the particular use. The U.S. Copyright Office has examined these issues and noted the development of voluntary licensing markets for AI training.

  Potential Policy Question

  Should federal policy provide creators with additional rights or mechanisms to know when—and determine how—their copyrighted works are used to train generative AI?

  Sources

  • GovInfo — S.2455, TRAIN Act
  • Congress.gov — H.R.8994, Protect Working Musicians Act of 2026
  • Congress.gov — S.1396, COPIED Act
  • Congress.gov — S.3813, CLEAR Act
  • U.S. Copyright Office — Part 3: Generative AI Training (Pre-Publication)