Tennessee has protected a person's voice, including simulations, as a property right since its ELVIS Act took effect on July 1, 2024. The law reaches more than a traditional endorsement dispute because it can cover an unauthorized simulated voice even when no original recording is copied.
That is only one state. American publicity rights still come from a patchwork of statutes and court-made rules, so the legal treatment of an AI-generated singer can change with the person involved, the place of the conduct, whether the performer is alive, and what the defendant actually did.
The gap matters in the musicians' identity claims against Suno because a multi-state case cannot assume every plaintiff owns the same right on identical terms. One jurisdiction may target commercial use of a voice, another may regulate digital replicas directly, and another may attach special rules to contracts or posthumous exploitation.
The law also creates potential liability beyond the person who publishes an imitation. It can reach someone who knowingly makes available a technology whose primary purpose or function is producing a particular identifiable person's photograph, voice, or likeness without authorization. That tool-focused provision is a much sharper lever than a rule limited to the final advertisement or song.
Its posthumous protection works differently again. Tennessee provides an initial ten-year period after death, but the right can continue when the identity keeps being commercially exploited. That makes duration depend partly on continued use rather than a single fixed expiration date.
California takes another route. Its digital-replica law for deceased personalities can impose liability when a highly realistic computer-generated voice or likeness appears without consent in an expressive audiovisual work or sound recording, while preserving specified exceptions for uses such as news, criticism, satire, parody, and some biographical material.
California also regulates certain contracts involving living performers. A clause authorizing a new digital performance can become unenforceable when it substitutes a replica for work the individual would have performed, lacks a reasonably specific description of intended uses, and was agreed without the representation specified by the statute.
That difference can matter before anyone compares how similar two synthetic voices sound. An estate may first need to establish that the deceased person fits the statute's territorial or domicile requirements, that the right still exists, and that the challenged conduct falls within the state's defined reach.
New York also distinguishes living people from deceased performers. Its long-standing rules prohibit certain unauthorized advertising or trade uses of a living person's name, portrait, picture, likeness, or voice, while its newer digital-replica provisions specifically address deceased performers and separate contract rules govern some replica clauses for living workers.
These are state digital-replica protections, not interchangeable versions of one national rule. California focuses heavily on deceased replicas and performer contracts, Tennessee expressly protects simulated voices and can reach certain tools, while New York combines advertising and trade protections with separate posthumous and contractual rules.
The reported federal bill would not simply stack another identical rule on top. Its preemption clause would generally displace state causes covering voice or likeness rights in digital replicas within expressive works, while preserving specified state causes that existed by January 2, 2025, along with certain sexual and election replica laws. That cutoff makes the age of a state rule potentially important.
Until that happens, a musician's strongest claim can depend on facts that barely appear in the generated audio itself. A living Tennessee singer confronting a simulated voice, a California estate confronting a replica in a new recording, and a New York estate protecting a deceased performer's identity may enter court through materially different statutory doors.
That difference also affects evidence. A case may turn on where the challenged use occurred, where a deceased performer was domiciled, whether a service merely hosted material or helped create it, whether consent covered a specific digital use, and whether the output qualifies as a statutory replica rather than simply evoking an artist's broader style.
That is only one state. American publicity rights still come from a patchwork of statutes and court-made rules, so the legal treatment of an AI-generated singer can change with the person involved, the place of the conduct, whether the performer is alive, and what the defendant actually did.
The gap matters in the musicians' identity claims against Suno because a multi-state case cannot assume every plaintiff owns the same right on identical terms. One jurisdiction may target commercial use of a voice, another may regulate digital replicas directly, and another may attach special rules to contracts or posthumous exploitation.
Tennessee reaches simulated voices directly
Tennessee's law is unusually explicit about voice. Its definition includes a sound in a recording or other medium that is readily identifiable and attributable to a particular individual, whether the sound contains the individual's actual voice or a simulation of it.The law also creates potential liability beyond the person who publishes an imitation. It can reach someone who knowingly makes available a technology whose primary purpose or function is producing a particular identifiable person's photograph, voice, or likeness without authorization. That tool-focused provision is a much sharper lever than a rule limited to the final advertisement or song.
Its posthumous protection works differently again. Tennessee provides an initial ten-year period after death, but the right can continue when the identity keeps being commercially exploited. That makes duration depend partly on continued use rather than a single fixed expiration date.
California takes another route. Its digital-replica law for deceased personalities can impose liability when a highly realistic computer-generated voice or likeness appears without consent in an expressive audiovisual work or sound recording, while preserving specified exceptions for uses such as news, criticism, satire, parody, and some biographical material.
California also regulates certain contracts involving living performers. A clause authorizing a new digital performance can become unenforceable when it substitutes a replica for work the individual would have performed, lacks a reasonably specific description of intended uses, and was agreed without the representation specified by the statute.
Location and death can change the available claim
The state split is not merely about damages. New York's posthumous publicity statute applies to a deceased personality who was domiciled in New York at death, and its protected right lasts forty years. California's corresponding statute instead contains seventy years and limits liability to acts occurring directly in California.That difference can matter before anyone compares how similar two synthetic voices sound. An estate may first need to establish that the deceased person fits the statute's territorial or domicile requirements, that the right still exists, and that the challenged conduct falls within the state's defined reach.
New York also distinguishes living people from deceased performers. Its long-standing rules prohibit certain unauthorized advertising or trade uses of a living person's name, portrait, picture, likeness, or voice, while its newer digital-replica provisions specifically address deceased performers and separate contract rules govern some replica clauses for living workers.
These are state digital-replica protections, not interchangeable versions of one national rule. California focuses heavily on deceased replicas and performer contracts, Tennessee expressly protects simulated voices and can reach certain tools, while New York combines advertising and trade protections with separate posthumous and contractual rules.
A federal proposal has not erased the patchwork
Congress has moved toward a national digital-replica right, but it had not enacted one by September 3, 2026. The current federal proposal was reported by the Senate Judiciary Committee in June and placed on the Senate legislative calendar, leaving state publicity law in force rather than replacing it with a completed national system.The reported federal bill would not simply stack another identical rule on top. Its preemption clause would generally displace state causes covering voice or likeness rights in digital replicas within expressive works, while preserving specified state causes that existed by January 2, 2025, along with certain sexual and election replica laws. That cutoff makes the age of a state rule potentially important.
Until that happens, a musician's strongest claim can depend on facts that barely appear in the generated audio itself. A living Tennessee singer confronting a simulated voice, a California estate confronting a replica in a new recording, and a New York estate protecting a deceased performer's identity may enter court through materially different statutory doors.
That difference also affects evidence. A case may turn on where the challenged use occurred, where a deceased performer was domiciled, whether a service merely hosted material or helped create it, whether consent covered a specific digital use, and whether the output qualifies as a statutory replica rather than simply evoking an artist's broader style.
