There’s something almost eerie about seeing an actor perform after they were never physically there to begin with. Not a posthumous cameo assembled from old footage, not a voiceover salvaged from archives—but a performance that exists because technology decided it could. That’s exactly what’s happening with Val Kilmer in As Deep as the Grave, and honestly, I can’t decide if it feels like a breakthrough or a quiet rupture in what we think acting even is.
When Presence Is No Longer Physical
The basic story is straightforward: Val Kilmer was cast in a role he never got to film due to serious illness. Years later, filmmakers used generative AI—with his family’s blessing—to place him into the movie anyway. On paper, that sounds like a respectful workaround. In practice, I think it forces us to confront a deeper question: what does it mean to “be in” a film?
Personally, I think we’ve long associated acting with physical presence—an actor’s body, voice, and spontaneity existing in a specific moment. What makes this case fascinating is that none of that happened here in the traditional sense. Instead, the performance is reconstructed from fragments: past images, voice models, and algorithmic interpolation. If you take a step back and think about it, we’re no longer watching a performance—we’re watching a simulation of intent.
And yet, I can’t dismiss it outright. Because intention still matters. Kilmer wanted to be part of the project. His family supported it. That complicates the ethical conversation in a way many critics tend to ignore.
Consent Changes Everything—Or Does It?
One thing that immediately stands out is how heavily the filmmakers emphasize consent. His children approved it. His estate was compensated. Industry guidelines were followed. On the surface, this looks like the “ideal” version of AI usage in film.
But in my opinion, consent alone doesn’t settle the debate—it just moves it into more nuanced territory. What many people don’t realize is that consent in creative work isn’t just legal; it’s philosophical. Did Kilmer consent to this exact performance, or to the idea of being included? There’s a difference, and it matters.
I find this particularly interesting because it introduces a new kind of authorship. Who is responsible for the performance? The director? The AI engineers? The editors? Or Kilmer himself, whose past likeness and voice form the raw material? What this really suggests is that we’re entering an era where authorship becomes distributed—and possibly diluted.
The Strange Poetry of Illness and Role
There’s another layer here that I find unexpectedly powerful. The character Kilmer “plays” suffers from tuberculosis, while Kilmer himself struggled with throat cancer. The filmmakers even leaned into this parallel, using his altered voice as part of the character.
From my perspective, this is where the project becomes more than just a tech experiment. It becomes symbolic. There’s something almost poetic—though slightly unsettling—about an actor’s real-life vulnerability being echoed through a digitally constructed performance.
At the same time, I wonder if we’re romanticizing something that is, at its core, مصنوع (constructed). Is this a meaningful artistic bridge between life and fiction, or are we projecting depth onto a technological workaround? Personally, I lean toward seeing it as both. That tension—between authenticity and fabrication—is exactly what makes this moment culturally significant.
Innovation or Economic Necessity?
Let’s be honest about another factor: this wasn’t just an artistic choice. It was a practical one. The production couldn’t afford reshoots. Recasting wasn’t feasible. AI became the solution.
And this is where I think the conversation gets uncomfortable. Because while the filmmakers frame this as creative innovation, it’s also about cost efficiency. If AI can replace the need for reshoots, scheduling, or even entire actors, what happens next?
In my opinion, this is the real story—not the emotional tribute, but the economic precedent. What starts as a respectful, one-off decision could easily evolve into standard industry practice. What makes this particularly fascinating is how quickly necessity turns into normalization in Hollywood.
The Legacy Question
Val Kilmer was already part of AI-assisted performance during his lifetime, notably recreating his voice for Top Gun: Maverick. That context matters. He wasn’t resistant to technology—he embraced it.
But here’s the deeper issue: legacy used to be something fixed. An actor’s body of work ended when they stopped working. Now, it doesn’t have to. Performances can be extended, expanded, even invented.
If you take a step back, this raises a profound question: who controls an artist’s legacy in the age of AI? Families? Studios? Algorithms trained on decades of footage?
Personally, I think we’re only beginning to understand how destabilizing this is. Because legacy used to be about memory. Now it’s becoming something closer to ongoing production.
A Glimpse of What’s Coming
What this really suggests is that Kilmer’s case isn’t an exception—it’s a preview. Today, it’s a respectful tribute backed by family consent. Tomorrow, it could be standard practice for studios to “revive” actors for new roles.
And I’ll be honest: I feel both impressed and uneasy. Impressed because the technology is undeniably powerful. Uneasy because it chips away at something we’ve always valued in performance—the idea that it’s fleeting, human, and tied to a specific moment in time.
If acting becomes something that can be generated, refined, and deployed on demand, then we’re no longer just watching stories—we’re watching engineered experiences. And maybe that’s where the real shift lies.
Not in whether this one film succeeds or fails, but in how quietly it redraws the boundaries of what it means to be present, to perform, and ultimately, to be remembered.
