AI: The Best Supporting Actor You’ll Never See

The Academy’s Neutrality Is a Choice

In a report from The Ankler, Erik Barmack argues the Academy has picked a carefully agnostic line: generative AI and other digital tools “neither help nor harm” a film’s Oscar chances, and branches will judge the work while considering “the degree to which a human was at the heart of the creative authorship.” (That same language appears in the Academy’s own rules update.)

The tension, Barmack noted, is that productions aren’t compelled to disclose AI workflows—and that silence becomes the de facto policy. Films can rack up nominations without voters being asked to factor in machine assistance; campaigns decide what to volunteer. It becomes a case of “don’t ask, don’t tell.”

The least controversial AI is the unsexy kind: dialogue cleanup, automated sound balancing, stabilization and other post-production (or “post”) efficiencies that save time without changing intent.

That’s AI as productivity—more output, fewer machine learning tools used for face replacement, facial performance modification and de-aging within an artist-driven workflow. It’s notable because “AI in films” often means machine learning inside the pipeline, quietly helping humans iterate.

Best (until it isn’t): AI Used to Fine-Tune Performances

“The Brutalist” became a flashpoint after an editor for the film noted that AI voice tech was used to fine-tune Hungarian pronunciations in post. Filmmakers emphasized it didn’t replace performances wholesale. This example is notable because once AI touches vocal or facial performances, audiences and voters start asking who authored the result.

“Emilia Pérez” is another example, where awards season coverage described AI blending vocals to broaden a performer’s singing range. The use of AI enhancement sits in the same moral neighborhood as auto-tune: accepted by many, but disclosure-sensitive.

Worst: Generative AI that Reads like a Shortcut

Marvel’s “Secret Invasion” drew backlash for AI-generated opening credits, with critics reading it as automation of artists rather than a creative necessity. The criticism highlighted reputational risk: audiences may tolerate invisible AI, but they react fast to “AI aesthetics.”

Indie horror film “Late Night With the Devil” also took heat following reports that it included AI-generated imagery, prompting clarifications from the filmmakers. Even limited generative imagery can become headline news.

The Ethical Third Rail: Synthetic Voices in Nonfiction

The Anthony Bourdain documentary “Roadrunner” drew criticism after it emerged that AI recreated Bourdain’s voice for a few lines, raising questions about consent and viewer trust. Since documentaries are billed as an authentic creation, synthetic audio can disrupt credibility.

Will the Oscars Telecast Include Any AI Applications?

There’s no public sign that the show will feature generative AI on air. But it’s hard to imagine a modern live broadcast without some machine learning in the plumbing, especially around accessibility and rapid clip distribution.

The Academy highlights live captions, live audio description and an ASL livestream on YouTube for the Oscars. And the captioning industry itself is evolving from purely human stenography toward advanced AI supported by human input. So even if the telecast never says “AI,” there’s a decent chance the technology is already helping viewers experience the show—just in the least glamorous ways imaginable.

For now, the Academy’s stance turns AI into Hollywood’s newest open secret: everywhere in the workflow, nowhere in the speech. The best uses so far keep humans at the center and the tech in the toolbox. The worst uses treat creativity like a cost center—or trust like an optional feature.

If AI wins anything this year, it’ll be Best Supporting Tool: always there, never thanked and absolutely not invited to the after-party.