Michael Caine is 93 years old. He has won two Academy Awards, anchored decades of British cinema, and delivered some of the most recognizable line readings in film history. Now his voice belongs, at least in part, to an artificial intelligence company. And it is already narrating Homer.
ElevenLabs, the AI audio firm that cloned Caine's voice with his permission, has released a 13-hour audiobook of The Odyssey featuring the actor's synthetic vocal likeness. The production, which The Hollywood Reporter detailed, includes a full cast of AI-generated voices, music, and sound effects. It reportedly took about six weeks to produce. It is available for free through ElevenLabs' ElevenReader app.
Six weeks. A 13-hour cinematic audiobook of one of Western civilization's foundational texts, narrated by a cloned version of one of the English-speaking world's most famous voices. Free of charge. That timeline alone tells you where the entertainment industry is headed, whether the guilds, the studios, or the regulators are ready or not.
Caine partnered with ElevenLabs last year to authorize the cloning of his voice. The arrangement covers not just The Odyssey but future books and articles as well, though the specific projects have not been disclosed. Financial terms of the deal remain undisclosed.
The actor framed the partnership in generous terms. In a statement, Caine said:
"The Odyssey is one of the greatest stories ever told. For nearly three millennia, its themes of perseverance, loyalty, temptation, and the enduring call of home have resonated across cultures and generations."
In earlier remarks reported when the partnership was first announced, Caine described ElevenLabs as "using innovation not to replace humanity, but to celebrate it." He added, as Breitbart reported, that the technology is "not about replacing voices; it's about amplifying them, opening doors for new storytellers everywhere."
That is a polished talking point. It may even be sincerely held. But it sidesteps the harder question: what happens to the working narrators, voice actors, and audiobook performers whose livelihoods depend on the very market that AI is now entering at zero cost?
ElevenLabs was founded in 2022. Its original purpose, AP News reported, was to dub audio into different languages while preserving a speaker's original voice and emotional tone. The company has since expanded aggressively into voice cloning and synthetic audio production.
That expansion has not been without controversy. In 2024, ElevenLabs' software was used to mimic President Biden's voice in a robocall sent to thousands of New Hampshire voters, a misuse that raised immediate alarm about the technology's potential for political deception. ElevenLabs now says it has measures in place to block the cloning of celebrity voices without consent.
The company also settled a lawsuit brought by actors Karissa Vacker and Mark Boyett, who accused ElevenLabs of using their voices without permission. The terms of that settlement, the jurisdiction, and the case details were not disclosed in available reporting. The fact that a settlement was reached, however, suggests the claims were serious enough that ElevenLabs preferred to resolve them quietly rather than litigate in public.
Questions about how AI tools intersect with influence, policy, and intellectual property are hardly confined to Hollywood. As we have reported, foreign influence operations have already targeted U.S. artificial intelligence policy, underscoring how high the stakes run when powerful technology outpaces the rules meant to govern it.
ElevenLabs has not limited itself to living performers. The company previously announced likeness rights deals with the estates of John Wayne, Judy Garland, Stan Lee, and Laurence Olivier. The implication is clear: if a voice was famous enough, ElevenLabs wants to license it, whether the owner is alive to consent or not.
There is a meaningful difference between a 93-year-old actor voluntarily signing a deal and an estate monetizing a dead person's vocal identity. But both paths lead to the same destination: a marketplace where human performance is optional and AI-generated content can be produced in weeks, distributed for free, and scaled without limit.
Matthew McConaughey is also involved with ElevenLabs, not just as a voice partner but as an investor, the New York Post reported. That dual role is worth noting. When a celebrity both licenses his voice and holds equity in the company profiting from the technology, his public endorsement of that technology carries a financial interest that audiences deserve to know about.
The timing of the Odyssey audiobook is not accidental. ElevenLabs posted on social media: "Before you see the film, hear the original epic." The film in question is Christopher Nolan's upcoming theatrical adaptation of The Odyssey, scheduled for release on July 17.
Nolan and Caine have a long professional history. The director cast Caine in the Dark Knight trilogy and Inception, among other films. ElevenLabs is plainly banking on that association to drive interest in its free audiobook, a marketing play that uses a legendary director's brand to promote an AI product the director had no apparent role in creating.
Whether Nolan views a synthetic Caine narrating Homer as a tribute or a distortion is an open question. The director has not commented publicly on the audiobook, at least not in any available reporting.
Strip away the press releases and the language about "amplifying" voices, and the commercial logic is straightforward. ElevenLabs produced a 13-hour, multi-voice, cinematic audiobook in roughly six weeks and released it for free. A traditional audiobook production of comparable scope, with a human narrator of Caine's stature, professional sound design, and studio time, would take months and cost a publisher a significant sum. The finished product would retail for $30 or more.
ElevenLabs did it in six weeks for nothing. The product is free. The message to the audiobook industry is unmistakable: your cost structure is about to collapse.
For consumers, free content narrated by a famous voice sounds like a gift. For the thousands of working voice actors, narrators, and audio engineers who depend on that market, it looks like something else entirely. The Screen Actors Guild spent months fighting over AI protections in its last contract cycle. This is exactly the scenario those negotiations were meant to address, and it arrived anyway, with the willing participation of one of the most celebrated actors alive.
Defenders of this arrangement will point out, correctly, that Caine consented. Nobody stole his voice. He signed a deal, issued a statement, and lent his name to the project. That distinguishes this case from the Vacker and Boyett lawsuit, where the actors alleged their voices were used without permission.
But consent from a single celebrity does not resolve the broader disruption. The audiobook narrators who will lose work to AI clones of Michael Caine did not consent. The mid-career voice actors competing against free, AI-generated content backed by venture capital did not consent. The market itself is being reshaped by a technology that treats human performance as raw material to be replicated and distributed at near-zero marginal cost.
ElevenLabs settled one lawsuit and signed a handful of celebrity deals. That does not make the company's business model safe, fair, or settled law. It means the company learned to get permission from people famous enough to sue, while the broader workforce watches the floor shift beneath it.
When a 93-year-old legend lends his voice to a machine and the machine produces a 13-hour epic in six weeks for free, the technology is no longer theoretical. The only question left is who pays the price, and it will not be the celebrities or the venture capitalists.
