Rights & Policy

Allowed works containing AI material while limiting awards to human creators who made qualifying creative contributions.

Defines covered musicians as humans and creates compensation and consultation rules for specified generative-AI soundtrack uses.

Would require visible and machine-readable disclosures for covered AI-generated audio and other media.

AFM alleges label AI licensing settlements breached collective-bargaining obligations concerning new uses and musician compensation.

Requires providers to use AI-transparency metadata when AI generated a material portion of artwork, composition, track, or music video; omission is treated as none.

Musician plaintiffs allege Mureka used copyrighted recordings without authorization for AI music training and generation.

BMG alleges Anthropic copied and used protected musical compositions and lyrics in connection with Claude models.

Bandcamp says it will not use or permit use of an artist's music to train music-generating models without express permission.

Bars music or audio generated wholly or substantially by AI and prohibits AI impersonation of artists or styles.

Defines cryptographically bound provenance manifests for media, adding Ogg Vorbis support while retaining other audio embedding mechanisms.

Expands postmortem protection against unauthorized digital replicas in sound recordings and audiovisual works, subject to exceptions.

Requires covered generative-AI developers to publish specified training-data documentation for systems made available to Californians.

Makes specified broad digital-replica clauses unenforceable when use is not reasonably specific and the performer lacked counsel or union representation.

Imposes provenance, disclosure, and detection-related duties on covered large generative-AI providers.

Music publishers allege copyright infringement involving song lyrics used with Anthropic's AI systems.

A second federal copyright action names Anthropic and individual defendants; the merits remain unresolved.

The court denied publishers' preliminary-injunction motion without resolving final liability.

Recommends a federal right against unauthorized digital replicas while leaving artistic style outside the proposed right.

Explains that assistive AI does not bar copyright, but protection reaches only sufficiently human-authored expression.

Analyzes training, fair use, licensing, and potential liability in a pre-publication report.

DDEX's AI working activity is developing music-supply-chain standardization for AI-related metadata and training permissions.

Directs federal review and challenge of certain state AI laws and seeks a national policy framework; it does not itself erase state statutes.

Certifies generative-AI products or models whose training data meets consent, open-license, public-domain, or ownership criteria plus diligence and recordkeeping.

Voluntary industry labeling program distinguishing AI-Generated from AI-Assisted sound recordings.

Articulates creator-industry positions favoring authorization, licensing, human authorship, transparency, and creator participation in AI policy.

Makes specified digital-replica contract provisions unenforceable unless intended uses are reasonably specific and representation conditions are met.

Adds digital-replica provisions to the Illinois Right of Publicity Act, including liability and exceptions for sound and audiovisual works.

Amends Illinois's digital-replica contract law to focus unenforceability on new performances and clarify consistent-use provisions.

Independent artists allege Suno infringed their sound recordings in training and outputs.

Musician plaintiffs allege Google's music-generation systems were trained on protected recordings without authorization.

Smith pleaded guilty to conspiracy to commit wire fraud and agreed to forfeiture; sentencing was set after the snapshot.

Introduces voluntary AI-Generated and AI-Assisted labels intended for sound-recording metadata and delivery systems.

Would have created a federal digital-replica right; it did not become law.

Proposes a federal right over unauthorized digital replicas with exceptions, remedies, and intermediary procedures.

Federal digital-replica bill reported favorably by the Senate Judiciary Committee; not enacted as of the snapshot.

Recommends a uniform federal AI framework that addresses intellectual property while preserving existing fair-use doctrine.

Roland and UMG principles emphasize human creativity, creator rights, transparency, and creator participation in responsible music AI.

Regulation (EU) 2024/1689 of the European Parliament and of the Council

Creates binding AI consent, compensation, notice, and dispute rules for covered labels and recording performers.

Requires opt-in consent before SoundCloud trains models intended to replicate a user's voice, music, or likeness and bars unauthorized impersonation.

Excludes recordings that are exclusively computer-generated, including by AI, from SoundCloud Monetization and distribution.

Combines impersonation enforcement, spam filtering, and support for granular AI disclosures through music metadata.

Allows compliant AI music but, beginning July 15, labels wholly AI-generated music and excludes it from royalty attribution.

Extends Tennessee personality rights to voice and creates liability pathways relevant to unauthorized AI voice replicas.

The D.C. Circuit affirmed that a work generated without human authorship is not copyrightable under the Copyright Act.

Requires applicants to disclose AI-generated material and claim only human-authored expression.

Opened the federal study that produced the Copyright Office's multipart AI report.

Advises that works lacking sufficient human authorship are ineligible for section 115 royalties and that suspicious claims may be investigated with royalties held in suspense.

Record-company plaintiffs allege unauthorized copying of sound recordings for model training and infringing outputs.

Record-company plaintiffs allege unauthorized copying for Udio's training and infringing outputs.

Federal prosecution of an AI-assisted music-streaming fraud scheme that used bot accounts and large volumes of generated tracks.

UMG and Udio announced resolution of their litigation and recorded-music and publishing licenses for a planned service.

WMG and Suno announced settlement of Warner's claims and a licensed partnership while other plaintiffs' claims continue.

WMG and Udio announced resolution of litigation and a framework for a licensed music-creation service.

Sets three platform principles: responsible AI collaboration with music partners, appropriate protections and opportunities, and scaled trust-and-safety enforcement.

Requires creators to disclose meaningful realistic altered or synthetic content, explicitly including synthetically generated music.

Expands likeness-detection tools that help eligible talent and representatives identify and request review of AI-altered likeness content.

Lets music partners designate fully, partly, or no generative-AI use through DDEX or CSV delivery metadata.

Allows privacy complaints seeking removal of realistic AI-generated or synthetic content resembling a person's face or voice.