Suspicious Minds: Elvis Presley's Rights Holders Take the ELVIS Act to Suno
Key Points
- The ELVIS Act gets an earlytest against a general-purpose music generator. The holder of Elvis's publicity rights joins the owners of 859 compositions in claims under Tennessee law, the Copyright Act and the DMCA.
- The plaintiffs claim Suno reproduces the voice itself, not an artist's “vibe.” They point to tracks allegedly generated from nothing more than a lyric that spells out his name.
- Suno's own words and deals are being turned against it. The complaint relies on Suno's statements in the Isbell class action and on its licensing arrangements with Warner Music Group, BMG and Believe.
Type a lyric that spells out E L V I S P R E S L E Y, the plaintiffs say, and Suno will sing it back to you in the King's voice.
That allegation is at the center of a suit filed Nov. 6, 2026, in the Northern District of California by the holders of Elvis Presley's publicity rights and music catalog against Suno, Inc., the developer of one of the most popular AI music generators.[1]
Two weeks earlier, many of the same rights holders sued Anthropic over lyrics. In my alert on that case, “All Shook Up — Elvis Presley's Rights Holders Sue Anthropic Over Lyrics Used to Train Claude,” I predicted that “the voice and likeness battles will likely be fought against audio and video generators.” It turns out I did not have to wait long. Thank you. Thank you very much.
While the Anthropic complaint was a copyright case about text, this one is a right of publicity case with a copyright case riding shotgun.
The Law Named for the King
There is a certain poetic justice in the ELVIS Act's first headline test involving Elvis. Tennessee named the Ensuring Likeness Voice and Image Security Act for the King, and, according to the complaint, Recording Academy officials working on the bill sought the blessing of Elvis's rights holders to use the name — something the complaint pointedly says Suno “does not know how to do.” In the Anthropic alert, I noted the irony that the Elvis-related suit did not invoke the Elvis-named statute. This one leads with it.
Several threshold questions will likely shape early motion practice:
- Choice of law. The case is pending in California against a Massachusetts-based company. The plaintiffs argue Tennessee law governs because Elvis was domiciled at Graceland when he, according to official reports, died in 1977, and Tennessee has the strongest interest in the outcome.
- Scope of the statute. Whether the ELVIS Act reaches a general-purpose music generator, as opposed to a tool marketed for cloning a particular singer, remains open. The complaint alleges that “[t]he primary purpose of Suno's technology and tools is to produce particular identifiable individual's voices.”
- Preemption and the First Amendment. Suno's Isbell motion previews a speech-based defense, and the overlap between the copyright and publicity claims may invite preemption arguments.
A Publicity Case First
Many of the plaintiffs, including Raleigh Music Publishing LLC, Abilene Music LLC, trusts associated with songwriter Luigi Creatore and the Aberbach family and Jason Simeon Aberbach, also brought the Anthropic suit. This time the group adds several individual rights holders and, most importantly, the holder of Elvis's publicity rights, and the complaint leads with that.
Two of the five counts arise under Tennessee law: violation of the ELVIS Act (Tenn. Code Ann. § 47-25-1101 et seq.) and the common law right of publicity. The other three are copyright claims covering 859 compositions: direct infringement (17 U.S.C. §§ 106, 501), circumvention of protection measures (§ 1201), and removal or alteration of copyright management information, or CMI (§ 1202).
The headline figure is the voice and likeness damages, “believed to exceed $50 million,” which the plaintiffs ask the court to treble. On top of that, they seek up to $128.85 million in statutory damages for copyright infringement (859 works at the $150,000 maximum for willful infringement), $2,147,500 for circumvention, and $21.475 million for CMI removal.
Suno has publicly maintained that it does not allow “any unauthorized use of an artist's work,” that it prohibits prompts using “specific artists' names and songs,” and that it works with third parties to screen for “potential misuse.”
The Name as ‘A Retrieval Key’
The heart of the complaint is the claim that Suno treats an artist's name as more than a word. According to the plaintiffs, Suno recognizes “Elvis” or “Elvis Presley” not as “a mere text string” but as “a retrieval key for a set of performer-specific representations.”
The plaintiffs allege that a “Voices” feature available to Pro account holders lets users save a vocal file and reuse it, and that one output is described as “Made from” a saved Voice titled, helpfully, “Elvis.” Most strikingly, they allege that, on Oct. 7, 2026, Suno produced two tracks titled “He Is Still The King,” each with a copy of Elvis's voice, from a prompt consisting of nothing but lyrics, including the line “I am rockin this rockabilly like E L V I S P R E S L E Y.” The original exhibits collected 50 recorded examples, including “Elvis in The Jailhouse Rock,” which allegedly had more than 125,000 plays. A First Amended Complaint filed Nov. 13, 2026, reportedly raises that number to 164, without changing the works or the claims.
If the “He Is Still the King” allegation holds up, that is significant. A name filter, however well designed, does not solve the problem if the model can find the voice through back channels. Spelling a name with spaces is not exactly a sophisticated jailbreak.
Suno's Own Words
Three days before the complaint was filed, Suno moved to dismiss a proposed class action brought by Jason Isbell and other musicians in the District of Massachusetts. As quoted in the complaint, Suno argued that its outputs “do not replicate the auditory characteristics of any real artist's voice to the point that someone would confuse a Suno creation with one actually sung by a particular famous musician,” and that Suno “has never allowed prompts for specific artists or copyrighted songs.”
The plaintiffs call those statements “unequivocally false” and offer their exhibits as “concrete proof of the exact opposite.” They also point to a user comment on one track: “I definitely thought Elvis was on this track.” Confusion evidence does not usually arrive so neatly packaged.
The timing is not an accident. Suno's Isbell defense rests in part on the argument that using an artist's name to describe a musical style is protected speech, and, according to press reports, Suno compared its technology to a producer asking musicians to create something with “more of a Jason Isbell vibe.” The Elvis plaintiffs' claim is not that Suno lets users ask for an Elvis vibe, but that Suno outputs the voice itself.
Headers, Footers and Copyright Management Information
Like the Anthropic complaint, this one focuses on how the data was acquired rather than on training alone. The plaintiffs allege that Suno used “stream ripping” tools to defeat YouTube's “rolling cipher” encryption, and used a third-party vendor's proxy and scraping services to get past login requirements, paywalls, CAPTCHAs, and bot defenses.
The CMI theory has a new twist. Rather than alleging a text-extraction tool chosen for its ability to strip credits, as in the Anthropic case, the plaintiffs allege that CMI typically sits in headers and footers, and that choosing to exclude them “is identical to making the deliberate choice to remove CMI from these files.” Whether that satisfies the “double scienter” standard of Stevens v. CoreLogic is likely to be contested. The plaintiffs also lean on Suno's admission, in its answer in the record label litigation, that its training data includes “essentially all music files of reasonable quality that are accessible on the open Internet,” making it “a near certainty” that the works were included.
Licensing Deals as Evidence
The complaint attempts to turn Suno's dealmaking into evidence. Suno settled with Warner Music Group, agreed to deprecate its then-current models, launched version 6 on Sept. 9, 2026, and has announced arrangements with BMG and Believe. The plaintiffs argue these deals show “a functioning market for licensing music and sound recordings for training data” and characterize Suno's strategy as “infringe now, make billions by being the first mover, and try to settle on the cheap later.” That is a fourth-factor fair use argument and a willfulness argument rolled into one. The complaint also highlights a July 31, 2026, ruling by the Munich Regional Court in favor of GEMA, Germany's music collecting society.
A Crowded Docket
Suno, like Anthropic, is collecting lawsuits. On the same day, publishers tied to Eminem's catalog filed a separate suit in the Northern District of California over 269 compositions, seeking damages that could exceed $40 million. That complaint highlights a track titled “Slim shady x Rosefalls” with vocals allegedly derived from Eminem's voice. According to the plaintiffs, the real Slim Shady does not have to actually stand up.
These cases join the original 2024 record label action, a second label suit filed Sept. 18, 2026, the Isbell class action, and suits by other publishers.
Practical Takeaways
- Know where the voice is protected. Post-mortem publicity rights vary considerably from state to state. Estates should confirm the decedent's domicile at death, which state laws apply, and exactly who holds the rights.
- Recorded outputs are now the evidence. This complaint relies on recorded outputs, play counts, user comments, and screenshots of search results. Documented testing of AI platforms is now table stakes.
- Filters must survive a spelled-out name. A guardrail that blocks “Elvis Presley” but not “E L V I S P R E S L E Y,” or that can be evaded with “in the style of,” invites exactly this kind of complaint. Test filters against workarounds and evaluate the model's ability to reproduce a specific voice directly.
- A commercial license is not a publicity clearance. A platform's grant of rights to exploit output does not necessarily protect a user from third-party publicity or copyright claims. Review vendor terms, indemnities, and exclusions before using AI-generated music commercially.
- Sound-alikes remain risky. As Midler v. Ford Motor Co. showed decades ago, imitating a famous voice can create liability even without copying a recording. AI makes imitation easier, not safer.
Whose Voice Is Singing
The Anthropic complaint asked where the lyrics came from. The Suno complaint asks whose voice is singing them.
The case is at its earliest stage. It has been assigned to Magistrate Judge Joseph C. Spero, Suno has not yet responded in court, none of the allegations have been tested, and Suno disputes them and has substantial defenses, including those it is already pressing in Isbell. But one thing is clear: the ELVIS Act was always going to get its day in court, and it was probably always going to involve Elvis.
Those developing, licensing, or deploying generative AI music tools should monitor this case closely, whatever the outcome.
[1] Complaint, No. 3:26-cv-13398 (N.D. Cal. filed Nov. 6, 2026); First Amended Complaint (filed Nov. 13, 2026). Quotations are from the original complaint.
For more information, please contact Anthony J. Noonan at 609.895.6621 or anoonan@foxrothschild.com or another member of Fox Rothschild's national Intellectual Property Department.
This information is intended to inform firm clients and friends about legal developments, including the decisions of courts and administrative bodies. Nothing in this alert should be construed as legal advice or a legal opinion. Readers should not act upon the information contained in this alert without seeking the advice of legal counsel. Views expressed are those of the authors and not necessarily this law firm or its clients.
