By Chris Cooke | Special Legal & Industry Report
Universal Music Group, Warner Music Group, American Federation of Musicians (AFM), Generative AI, Intellectual Property Law
Executive Overview
As the global music industry rushes headlong into a gold rush of multi-million-dollar generative artificial intelligence licensing deals, a foundational legal battle is brewing in the United States that could rewrite the economic model of recorded music. At the heart of the controversy is a high-stakes lawsuit filed by the American Federation of Musicians (AFM)—one of the nation’s most prominent and powerful labor unions representing instrumentalists—against two of the world’s three major music conglomerates: Universal Music Group (UMG) and Warner Music Group (WMG).
The core of the dispute centers on how legacy union contracts apply to cutting-edge technology. Specifically, the AFM claims that Universal and Warner have breached their contractual obligations under the Sound Recording Labor Agreement (SRLA) by failing to compensate, or even consult, session musicians whose recorded performances were allegedly utilized to train generative AI audio models. These models are built by tech platforms such as Udio and Suno, both of which have recently inked various partnerships or found themselves embroiled in broader legal reckonings with major rightsholders.
As autumn court filings unfold, the battle lines have hardened. The major record labels are aggressively seeking to have the lawsuit dismissed, arguing that the contract’s terminology is unambiguous and that the union is attempting to stretch outdated clauses to cover technological phenomena that were unimaginable when the pacts were penned. Conversely, the AFM insists that the contract’s "new use" clauses were specifically engineered precisely to protect musicians when their existing recorded works are repurposed for novel, unanticipated commercial applications.
This legal collision is more than a mere contract dispute; it is a bellwether for the entire creative economy. As generative AI threatens to disintermediate traditional session players—training models on their past sweat and skill only to generate competing, synthetic audio files—the outcome of this lawsuit will establish vital precedents for labor rights, compensation models, and the limits of copyright law in the age of machine learning.
Detailed Chronology: From AI Pacts to Federal Court
To understand how the music industry arrived at this critical juncture, it is necessary to retrace the timeline of both technological disruption and legal escalation.
Spring to June 2024: The Lawsuit Takes Shape
The tension spilled into the public domain in June 2024, when the AFM formally filed its landmark lawsuit against Universal Music Group and Warner Music Group in a U.S. federal court. The union’s complaint alleged a systematic breach of the Sound Recording Labor Agreement (SRLA). According to the AFM, the major labels entered into a series of generative AI licensing agreements—notably involving platforms like Udio (for Universal) and both Udio and Suno (for Warner)—without fulfilling the mandatory notice and compensation requirements owed to the session musicians who performed on the master recordings utilized in those training pipelines.
For months following the initial filing, the legal teams for both major labels prepared their defenses, signaling early on that they would move for an outright dismissal of the complaint. They framed the lawsuit as a legally meritless attempt by a labor organization to hijack commercial technology agreements that fall entirely outside the scope of traditional collective bargaining parameters.
Autumn 2024: The Motion to Dismiss and New Filings
The dispute escalated significantly this month as both sides submitted comprehensive new filings to the court in response to the labels’ motions to dismiss.
- The Majors’ Position: Universal and Warner doubled down on their stance that the disputed language within the SRLA is entirely clear, narrow, and fundamentally incompatible with the AFM’s demands. Universal argued in its latest filing that the contract’s "new use" clause is "clear and unambiguous" in its favor, asserting that the union has failed to mount a coherent legal counterargument, instead offering a "scattershot series of arguments that obscure rather than confront what the contract makes clear."
- The Union’s Position: The AFM fired back, refusing to back down. The union argued that the text of the SRLA imposes an absolute, mandatory payment obligation whenever a signatory record label puts a protected, covered sound recording to a purpose not explicitly covered by the base agreement. Furthermore, while the labels claim that the absence of a dedicated, pre-existing "AI agreement" invalidates the union’s claims, the AFM maintains that existing mechanisms within the contract—such as baseline session rates, streaming calculations, and sampling formulas—provide more than enough objective measures for a court to calculate statutory damages.
Supporting Context & Metrics: The Anatomy of the "New Use" Clause
At the center of this legal chess match is a dense, highly specialized contractual mechanism known in the trade as the "new use" clause of the Sound Recording Labor Agreement (SRLA). To appreciate the gravity of the arguments being waged in federal court, one must examine the mechanics of this clause and the broader economic realities facing modern session musicians.
What is the SRLA and the "New Use" Clause?
The Sound Recording Labor Agreement is the bedrock collective bargaining framework governing the employment of session instrumentalists, arrangers, orchestrators, and copyists working on commercial sound recordings in the United States. It dictates everything from minimum hourly wages and health and pension contributions to rules governing how master recordings can be exploited.
Embedded within this agreement is the "new use" clause. Historically, this clause was designed to protect musicians from technological shifts—such as the transition from vinyl to cassette, the rise of sync licensing in film and television, or the commercial deployment of digital sampling. The spirit and letter of the rule dictate that:
- If a record company uses an existing sound recording in a new way that was neither anticipated nor explicitly covered by the baseline terms of the SRLA at the time of recording, an obligation is triggered.
- The label is contractually required to compensate the musicians who performed on that recording.
- The label is required to formally notify the union of this new commercial deployment.
The Core Interpretive Divide
The legal battle hinges entirely on how both parties interpret the conditional triggers of this clause:
[Master Recording Created Under SRLA]
│
▼
[Label Licenses Recording for AI Training] ──> Is this a "New Use"?
│
├──────────────────────────────────────────┐
▼ ▼
[MAJOR LABELS' ARGUMENT] [UNION'S (AFM) ARGUMENT]
• Clause requires a separate, • Clause is a broad, mandatory
pre-existing union agreement payment obligation for *any*
governing the specific new use. uncovered purpose.
• No specific "AI agreement" exists. • Existing SRLA metrics (session,
• Therefore, the clause does not apply. streaming, sampling) provide damages.The Labels’ Defense: The Pre-Requisite Agreement Theory
Universal and Warner argue that the text of the "new use" clause does not operate in a vacuum. Specifically, the clause dictates that when a recording is utilized for a new purpose, the musician must be paid an amount equal to what they would have received if they had been brought back into a studio to record music specifically for that new purpose, based on whatever standard union agreement governs that specific type of recording session.
Because generative artificial intelligence training and synthesis represent entirely unprecedented technological paradigms, no specialized "AI recording agreement" currently exists within the AFM’s catalog of collective bargaining pacts. Therefore, the majors argue, because there is no separate agreement to import rates from, the "new use" clause is structurally incapable of being triggered in this context. Universal stressed in its filings that the clause "does not state a rate," but rather "imports one from the separate AFM agreement governing the new purpose"—an agreement that, by all admissions, does not yet exist.
The Union’s Rebuttal: Broad Protections and Objective Damages
The AFM views this defense as a clever semantic trap designed to strip musicians of foundational protections during a technological revolution. The union counters that:
- The text of the SRLA imposes a broad, mandatory payment obligation the moment a signatory company diverts a covered recording to any purpose not explicitly covered by the agreement.
- The absence of a bespoke "AI agreement" does not grant labels a free pass to exploit master recordings royalty-free.
- Existing, time-tested provisions within the SRLA—including baseline session rates, digital streaming multipliers, and commercial sampling formulas—supply perfectly adequate, objective measures from which a court can calculate financial damages.
Furthermore, the AFM has adopted a more flexible legal posture regarding the contractual ambiguity itself. Under civil procedure rules, if a plaintiff can convince a federal judge that a disputed contract term is "reasonably susceptible to more than one interpretation" and that the plaintiff’s interpretation is at the very least "plausible," the motion to dismiss must be denied, allowing the case to proceed to discovery.
Official Statements & Industry Perspectives
The litigation has polarized the broader entertainment and labor ecosystems, drawing sharp contrasts between corporate rightsholders and creative labor advocates.
- Universal Music Group Legal Team: In recent court submissions, UMG characterized the union’s litigation strategy as a reactive scramble, stating that the AFM has made "a scattershot series of arguments that obscure rather than confront what the contract makes clear." UMG maintains that the disputed text is exceptionally straightforward, leaving zero room for the expansive liabilities the union attempts to impose.
- The American Federation of Musicians (AFM): In its public statements and formal legal filings, union leadership has remained unyielding. The AFM asserts that major labels cannot selectively embrace generative AI as a lucrative new revenue stream while simultaneously attempting to bypass the historical labor protections embedded in their collective bargaining agreements. The union argues that allowing labels to mine master recordings for AI training without compensation undermines the entire fabric of session work.
- Wider Music Industry Observers: Independent legal analysts note that the case highlights a profound systemic blind spot in modern music business deals. While major labels, music publishers, and independent distributors have rushed to strike multi-million-dollar data-licensing pacts with tech giants like OpenAI, Google, Suno, and Udio, session musicians have been conspicuously absent from the revenue-sharing table. While top-tier superstar artists may negotiate individual consent and payout terms, the invisible workforce of session players—whose instrumental tracks form the DNA of countless training datasets—rely almost exclusively on collective bargaining and legacy labor agreements to secure a stake in emerging digital economies.
Future Outlook: What This Means for the Music Industry
As the federal court weighs the motions to dismiss filed by Universal Music Group and Warner Music Group, the broader implications of this lawsuit extend far beyond the immediate litigants.
1. Immediate Judicial Roadmap
If the federal judge sides with the major labels and grants the motion to dismiss, the AFM will face a severe roadblock in utilizing legacy collective bargaining agreements to police generative AI training. Such a ruling would essentially force labor unions to abandon retrospective contract enforcement and rely entirely on future, painstaking rounds of collective bargaining to secure AI-specific protections.
Conversely, if the judge denies the motion to dismiss—ruling that the union’s interpretation of the "new use" clause is plausible—the case will advance to the discovery phase. This would compel UMG and Warner to open their internal communications, data-licensing contracts with AI firms, and financial ledgers to legal scrutiny, potentially unearthing explosive details regarding how master recordings were transferred to tech developers.
2. The Intersection of Labor Negotiations and Copyright Reform
Regardless of the immediate courtroom outcome, the dispute underscores a harsh reality for modern performers: relying solely on decades-old collective bargaining agreements to govern transformative technologies is an inherently precarious strategy.
Moving forward, fair compensation for session musicians in the age of generative AI will likely require a two-pronged approach:
- Aggressive Union Modernization: Future iterations of the SRLA and comparable union pacts across SAG-AFTRA and other creative guilds must explicitly incorporate precise definitions, consent frameworks, and mandatory remuneration schedules for machine learning, model training, and synthetic audio generation.
- Copyright Law Reform: Legislative intervention at the federal level—clarifying fair use exemptions, establishing strict data transparency mandates for AI developers, and guaranteeing moral rights and compulsory licensing for performers—will be essential to prevent algorithms from legally cannibalizing human artistry.
For now, all eyes remain fixed on the federal bench. The judge’s decision on whether to let the AFM’s lawsuit proceed will serve as an early, decisive blueprint for how the law intends to balance corporate technological innovation against the fundamental rights of the musicians who built the recorded music canon.
