• 2023: Hollywood’s AI alarm bells start ringing
• 2024–2025: AI actors, deepfake stars, and growing unease
• May 1–2, 2026: The Academy makes its move
• The humanist perspective: Protect the craft, protect consent
• The technophile critique: Yesterday’s rules for tomorrow’s art?
• The institutional balancing act: Norm‑setting in real time
• 2027 and beyond: Oscars as a cultural firewall

Academy of Motion Picture Arts and Sciences Rules AI-Generated Performances and Scripts Ineligible for Oscars Human Human coverage emphasizes that the Academy’s rule change is designed to ensure only consented, demonstrably human performances and human-authored scripts are eligible for Oscars, largely as a response to labor concerns raised in recent Hollywood strikes. It portrays the policy as reinforcing human creative primacy and protecting actors’ likenesses and writers’ work, while still allowing AI tools in production so long as humans remain the core creative agents. @Verge @TC The Oscars have drawn a bright red line in the sand: AI can help make movies, but it can’t win the gold. With new rules for the 99th Academy Awards, the film industry’s most powerful arbiter has decided that when it comes to acting and writing trophies, silicon brains are out and human hands — and faces — are in.

§2023: Hollywood’s AI alarm bells start ringing

The road to the Academy’s ruling runs straight through the Hollywood labor wars of 2023. As generative models exploded in capability, actors and writers realized their own faces and words might soon be cloned, remixed, and monetized — possibly without their consent.

AI became “one of the main sticking points in the actors’ and writers’ strikes back in 2023,” as unions fought to keep studios from using machine‑generated dialogue and digital doubles as a cheap replacement for human talent.¹ The message from picket lines was blunt: AI could be a tool, but it could not be allowed to become a scab.

Those strikes didn’t directly change Oscar rules, but they changed the climate. What had been a theoretical problem became a contractual one. Once unions started hard‑coding protections around AI use, it was only a matter of time before award bodies felt pressure to answer a basic question: if a performance or script is partially or wholly generated by AI, who — or what — are you actually honoring?

§2024–2025: AI actors, deepfake stars, and growing unease

As the tech improved, the examples got more concrete — and more uncomfortable.

An independent film moved forward with an AI‑generated version of Val Kilmer, raising eyebrows about how far “digital resurrection” of performers might go.¹ Meanwhile, AI “actress” Tilly Norwood — a fully synthetic, generative creation — kept popping up in headlines, a kind of stress test for how audiences and institutions would treat non‑human “stars.”¹

New video models pushed things further, enabling realistic human faces and bodies that could be conjured from prompts or stitched together from training data. The reaction inside parts of the industry was grim; some filmmakers were already making “sweeping declarations of despair” about AI’s impact on their craft and careers.¹

Outside Hollywood, adjacent creative fields were reaching their own breaking points. At least one novel was pulled by its publisher over apparent AI use in its creation, and writers’ groups began declaring that works using generative tools would be ineligible for their awards.¹ Step by step, a norm was emerging: if an honor is meant to celebrate human authorship, AI involvement has to be tightly circumscribed — or outright banned.

§May 1–2, 2026: The Academy makes its move

Against that backdrop, the Academy of Motion Picture Arts and Sciences formally updated its rulebook for the 99th Academy Awards, which will air in 2027. The headline policy could not be clearer: “The organization behind the Oscars says that only humans can get acting awards.”²

Under the new rules, “only roles credited in the film’s legal billing and demonstrably performed by humans with their consent will be considered eligible” for acting Oscars.² In other words, if your “actor” is a purely AI‑generated avatar like Tilly Norwood — or a composite face synthesized in post — it can’t be nominated.

The Academy extends that logic to the page as well as the screen. Screenplays eligible for writing awards must be “human‑authored,” a phrase that draws a bright line against scripts spit out wholesale by large language models.¹ That dovetails with the wider backlash in publishing and literary circles, and serves as a direct response to the fear that studios could use AI to churn out first drafts, then bring in human writers merely as polishers.

Anticipating gray areas, the Academy has also given itself investigative powers. If “questions arise” about generative AI use in a film, it “can ‘request more information about the nature of the use and human authorship.’”² Another formulation of the same stance: the Academy “said it has the right to request more information about a film’s AI usage and ‘human authorship.’”¹

In regulatory terms, that’s a classic “trust but verify” posture. The Academy is not banning AI from film production outright; it is reserving the right to look under the hood when a performance or script looks suspiciously inhuman.

§The humanist perspective: Protect the craft, protect consent

From a human‑first point of view, the new rules are less an act of conservatism than an overdue defense of what Oscars were supposed to reward in the first place.

The basic standard — that acting awards are only for roles “demonstrably performed by humans with their consent”² — echoes union demands from the 2023 strikes and responds directly to anxieties about digital replicas. If an actor’s scanned likeness can be reused indefinitely, the concept of a single, time‑bound “performance” starts to dissolve. By tying eligibility to consented, documented human work, the Academy is pulling performance back into the realm of labor rather than data.

Similarly, requiring that screenplays be “human‑authored”¹ is a statement that the Oscars’ writing categories are about original human expression, not about who can engineer the cleverest prompt. It also signals solidarity with writers in other media who have begun stripping AI‑assisted works of award eligibility.¹

This camp sees the Academy’s move as aligning the Oscars with their own mythology: the statue is meant to honor artists who sweat for their craft, not the servers that processed their training data.

§The technophile critique: Yesterday’s rules for tomorrow’s art?

On the other side are technophiles and some avant‑garde creators who argue that the Academy is drawing its boundary too crudely — and too early.

They would point to projects like the independent film using an AI‑generated Val Kilmer as evidence that the line between “AI” and “human” performance is already blurred.¹ If a living actor collaborates with an AI system, directing its movements, providing reference material, and curating outputs, is the final on‑screen presence really non‑human? Or is it a new kind of co‑authored performance that the rules simply don’t know how to classify yet?

The same critique applies to screenplays. Many writers already use AI as a brainstorming partner, for structure help, or to beat back writer’s block. The Academy’s “human‑authored” requirement sounds straightforward — but what does it mean in practice when virtually any writing software can embed generative suggestions? Without a clear threshold, technophiles worry that the policy could chill innovation or push filmmakers to conceal their AI use rather than experiment openly.

There’s also a broader cultural anxiety here: as “AI‑generated actors and scripts are now ineligible for Oscars,”¹ some fear the Academy is cementing a binary that will rapidly become obsolete, much like early bans on digital color correction or motion capture once looked fussy and shortsighted.

§The institutional balancing act: Norm‑setting in real time

From the Academy’s vantage point, however, this is less about freezing the future than buying time.

The rules arrive while AI “actress” Tilly Norwood “keeps making headlines,” while new video models are rattling directors, and while other award‑granting bodies are scrambling to catch up.¹ If the Oscars did nothing, they risked becoming the first major institution to hand a golden statue to a fully synthetic performer — or to a script typed by a chatbot.

Instead, they’ve done three things at once:

1. Affirmed a principle: Only human beings can be recognized as actors and screenwriters by the Academy.²
2. Set a consent standard: Performances must be “demonstrably performed by humans with their consent,” not stitched together in a dark data lab.²
3. Reserved flexibility: The Academy “can request more information about the nature of the use and human authorship” when AI is involved, giving itself room to refine the line later.²

That last point may prove the most important. It acknowledges that AI is already woven into filmmaking and that the institution will need more nuance as hybrid practices emerge.

§2027 and beyond: Oscars as a cultural firewall

When the 99th Academy Awards air in 2027, the winners’ list will double as a cultural statement: these are, by rule, human achievements. Behind the scenes, though, many of those films will almost certainly have been touched by AI tools for editing, effects, or even early drafts of dialogue.

The Academy’s new rules don’t eliminate AI from Hollywood. They do something subtler but symbolically potent: they insist that at the top of the pyramid — where names go on statues and history books — the credit still belongs to people.

Whether that stance looks principled or reactionary in ten years will depend on how fast the line between human and machine creativity blurs. For now, the Oscars have chosen their side.

§1. TechCrunch — “AI-generated actors and scripts are now ineligible for Oscars.”

§2. The Verge — “Only roles credited in the film’s legal billing and demonstrably performed by humans with their consent will be considered eligible.”

Story coverage