Legal Battle Intensifies Between Musicians Union AFM and Major Music Labels Over Generative AI Licensing Deals

As the global music industry rushes headlong into a lucrative yet deeply controversial era of artificial intelligence partnerships, a high-stakes legal collision is unfolding in the United States. The American Federation of Musicians (AFM), one of the most prominent labor organizations representing instrumentalists and performers in North America, finds itself locked in a fierce courtroom battle with two of the world’s most powerful major record labels: Universal Music Group and Warner Music Group. At the heart of this legal dispute is a fundamental disagreement over contractual obligations, specifically how decades-old collective bargaining agreements apply to the cutting-edge technology of generative AI training models.

While music executives tout multi-million-dollar licensing agreements with artificial intelligence startups as a necessary step toward monetizing technological innovation and protecting copyright integrity, labor advocates see a systematic evasion of artist compensation. The lawsuit, initially filed earlier this year, targets major licensing arrangements that the labels struck with generative AI firms such as Udio and Suno. As both sides fire fresh legal salvos in court, the outcome of this litigation could permanently alter the economic landscape of recorded music, setting a monumental precedent for how session musicians and background performers are compensated in the age of machine learning.

Background Context and the Origin of the Dispute

The genesis of this legal confrontation traces back to the rapid commercialization of generative artificial intelligence platforms capable of producing convincing musical compositions and vocals from scratch. These models require massive datasets to train on—vast repositories of existing audio recordings that teach algorithms melody, rhythm, instrumentation, and harmony. Unsurprisingly, major record labels, holding vast vaults of copyrighted master recordings, emerged as prime partners for AI companies seeking legitimate training data.

However, the rapid deployment of these deals bypassed a crucial constituency within the music ecosystem: the musicians, session players, and instrumentalists who physically performed on the tracks utilized to train these algorithms. While star recording artists often negotiate custom, high-level payouts when their labels cut enterprise-level deals, anonymous session musicians whose contributions form the backbone of these master recordings are frequently left out of the financial equation.

Enter the American Federation of Musicians. Tasked with safeguarding the financial security and professional rights of its members, the union stepped forward to challenge the majors. The legal vehicle for this challenge is the Sound Recording Labor Agreement (SRLA), a comprehensive collective bargaining framework that governs the employment terms, working conditions, and royalty distributions for musicians hired by major labels in the United States.

The Core Legal Contention: Decoding the New Use Clause

The entire lawsuit hinges on the judicial interpretation of a specific contractual provision within the SRLA known as the "new use clause." Drafted long before the advent of artificial intelligence, this clause was designed to protect performers when master recordings are repurposed in ways not originally anticipated or explicitly covered under standard studio recording contracts.

Under the traditional parameters of the new use clause, if a record label takes an existing sound recording and applies it to a novel commercial medium or purpose—such as synching it to a new format or repurposing it outside its original scope—the label incurs a mandatory legal obligation. Specifically, the company must compensate every musician who appeared on that original recording as if they had been brought back into a studio specifically to record music for that new purpose. Furthermore, the label is legally required to formally notify the union of this new application.

To date, neither Universal Music nor Warner Music has fulfilled either of these mandates in connection with their respective AI licensing agreements. The labels have consciously withheld notification and compensation to AFM members whose performances were fed into generative AI models.

The Legal Arguments: Textual Rigor Versus Contractual Ambiguity

The legal sparring playing out in federal court centers on a stark divergence in contractual interpretation. The defense mounted by Universal Music and Warner Music rests on a strict, text-based reading of the SRLA’s new use clause.

In their recent court filings aimed at securing an outright dismissal of the lawsuit, the major labels argue that the new use clause is entirely unambiguous and inherently favors their position. Their core defense is built on a technicality within the clause’s compensation mechanism. The SRLA dictates that when a recording is deployed for a new use, the compensation owed to the musician must match what would be required under the specific AFM agreement governing that exact type of recording session.

Because generative AI is a revolutionary technology, no separate, dedicated AFM labor agreement currently exists to govern AI training sessions or AI-generated outputs. Consequently, the majors argue, because there is no parallel union agreement outlining AI session rates, the new use clause fundamentally cannot be triggered. In Universal Music’s framing, the clause "imports" a rate from a non-existent agreement, rendering the provision inapplicable to the current dispute. The labels maintain that AFM itself conceded during preliminary discussions that no such specialized AI agreement exists, thereby pulling the rug out from under its own legal theory.

Conversely, the American Federation of Musicians views the situation through an entirely different legal lens. AFM rejects the notion that a separate, pre-existing AI agreement is a prerequisite for the new use clause to apply. In its counter-filings, the union insists that the text of the SRLA imposes a clear, mandatory payment obligation the moment a signatory record company consigns a covered recording to a purpose not explicitly covered by the baseline agreement.

Addressing the labels’ argument regarding the absence of a designated AI rate schedule, AFM asserts that the court does not need a brand-new, purpose-built AI agreement to calculate damages. The union argues that existing provisions within the broader SRLA—including established session rates, streaming royalty scales, and digital sampling rates—provide robust, objective financial metrics from which a court can readily determine fair compensation and statutory damages.

Crucially, while the major labels are desperate to convince the presiding judge that the contract is crystal clear in their favor—which would allow the judge to throw the case out immediately—AFM has adopted a more flexible litigation strategy. The union freely concedes that the new use clause may be "reasonably susceptible to more than one interpretation." Under standard contract law, if a judge finds that a contractual provision is ambiguous and that both sides have presented plausible interpretations, the case cannot be dismissed prematurely. Instead, it must proceed to the discovery and trial phases to determine the true intent of the contracting parties. AFM maintains that its reading of the clause is, at the bare minimum, entirely plausible, thereby barring the labels’ motion for dismissal.

Timeline of Events

• June 2024: The American Federation of Musicians formally files a multi-count lawsuit in federal court against Universal Music Group and Warner Music Group, alleging widespread breach of the Sound Recording Labor Agreement regarding unauthorized AI licensing deals with firms like Udio and Suno.
• Late Summer / Early Autumn 2024: Legal teams for Universal and Warner formulate motions to dismiss the lawsuit, arguing that the SRLA’s new use clause does not apply to generative artificial intelligence training due to the absence of specific industry-wide AI wage scales.
• November 2024: Both parties submit comprehensive new legal filings. The major labels double down on claims that the contract is unambiguous and favors dismissal, while AFM counters by asserting that existing contractual rates can serve as valid baselines for damages, urging the court to reject the dismissal bid.

Broader Industry Impact and Future Implications

The ripple effects of this litigation extend far beyond the balance sheets of Universal Music and Warner Music. As the major record labels continue to announce lucrative, high-profile partnerships with technology firms to capitalize on the generative AI boom, the question of equitable creator compensation remains the industry’s most contentious ethical and legal frontier.

While top-tier recording artists frequently command individual legal representation and lucrative carve-outs in their personal label contracts, session musicians, background instrumentalists, and orchestral players rely almost exclusively on collective bargaining agreements like the SRLA to secure their livelihoods. If major record companies can successfully harvest decades of recorded performances to train commercial AI models—which can subsequently generate competing audio content—without compensating the original human creators, the economic foundation of session work could be severely undermined.

Furthermore, this case highlights a broader regulatory and legal vacuum. As copyright laws struggle to keep pace with rapid advancements in machine learning, labor unions are stepping into the breach to protect workers’ rights through contract enforcement. If AFM succeeds in forcing the majors to recognize AI training as a "new use" under legacy agreements, it will establish a powerful legal precedent. Such a ruling would compel record labels to renegotiate how technical data sets are licensed and ensure that the human talent responsible for creating the foundational art is neither replaced nor economically disenfranchised by the very technology trained on their sweat and skill.