Examples of AI in Movies (2026)

AI in movies shows up as de-aging VFX, voice cloning, AI-assisted rotoscoping, and studio equity stakes in AI companies — not as fully autonomous production. The examples below are grounded in verified production details from major Hollywood films and series through mid-2026, covering what AI actually did, which vendors were involved, and what the consent and union agreements looked like.

De-Aging and Digital Face Work

Indiana Jones and the Dial of Destiny (2023) is probably the most technically ambitious de-aging project completed to date. ILM built a proprietary tool called ILM FaceSwap — combining CGI, machine learning, and photogrammetry — to de-age Harrison Ford by roughly 35 years for the film's 25-minute WWII prologue. The team trained the system on hundreds of hours of archival Ford footage from the original trilogy. Over 100 artists worked on it for three years.

Here (2024) took a different approach. Director Robert Zemeckis used Metaphysic's deepfake technology directly on set, in real time, so he could see the de-aged final look on Tom Hanks and Robin Wright while they performed — without heavy prosthetics or post-production VFX. Metaphysic won the VES Emerging Technology Award in 2025 for the work.

Alien: Romulus (2024) used AI to recreate the character Ash after original actor Ian Holm died in 2020. The production required estate consent and compensation before proceeding — a pattern that's now standard for AI likeness work.

Smaller-scale but widely used: Vanity AI, a digital touch-up tool that handles skin texture and blemish correction frame-by-frame. It was used on Kathryn Newton's Cassie Lang in Ant-Man and the Wasp: Quantumania (2023), and reportedly on Spider-Man: No Way Home before that. For a broader look at how these tools fit the AI filmmaking workflow, the same machine-learning principles apply across both studio and independent productions.

For the Bob Dylan biopic A Complete Unknown (2025), the production used a tool called Revize to replace a stuntman's face with Timothée Chalamet's in select action shots — a narrow, specific application that kept the work contained and cost-effective.

AI Voice Cloning

Lucasfilm's voice work for Obi-Wan Kenobi (Disney+, 2022) set a precedent that's still being cited. Skywalker Sound partnered with Respeecher, a Ukrainian AI voice startup, to recreate a younger version of James Earl Jones's Darth Vader voice. The system was trained on archival recordings of Jones from earlier decades. Jones, then 91, reviewed and approved the output — and later signed over rights for Lucasfilm to use his AI-modeled voice in future projects.

That consent-first framework matters. The SAG-AFTRA union has pushed for the NO FAKES Act through 2025–2026, which would require explicit consent before any AI-generated likeness or voice is used commercially. The Jones arrangement is frequently cited as the model other productions should follow.

ElevenLabs released a "Video-to-Sound" tool in 2025 that analyzes VFX footage and auto-generates foley and sound effects to match on-screen action. Productions have used it in post-production to fill in audio for VFX shots — a task that traditionally requires a Foley artist and ADR sessions. The broader question of AI voices vs voice actors is worth understanding separately from the consent and legal layer covered above.

Studio-Level AI Partnerships

Two deals announced in mid-2026 signal that AI integration is moving from production tools to ownership stakes.

Lionsgate and Runway (2024–2026): In 2024, Lionsgate became the first major Hollywood studio to formally partner with an AI firm — Runway — for production tooling. By June 2026, Lionsgate had taken an equity stake in Runway and announced plans for AI-generated short-form episodic series based on existing IP.

Google DeepMind and A24 (June 2026): Google announced a $75 million equity investment in A24 on June 22, 2026 — Google's first direct equity stake in a Hollywood studio. A24 Labs, a 20-person internal unit, is building AI-assisted production tools: storyboard generators that take scripts as input, scheduling tools, and related pre-production software. DeepMind researchers are embedded inside active A24 productions. Notably, the deal explicitly does not give Google access to A24's content library or training data — a detail A24 negotiated to protect its IP.

Workflow and Post-Production Tools

Rotoscoping — isolating subjects from backgrounds frame by frame — is one of the most labor-intensive tasks in post. On Everything Everywhere All at Once, editors used Runway's AI rotoscoping to remove crew members and pulley rigs from shots in the film's alternate-universe climax sequence. Work that might take weeks was turned around in days.

AI dubbing and localization is another area seeing rapid adoption. Localization timelines have compressed significantly with AI lip-sync dubbing tools, according to industry reporting from 2025. The trade-off — voice quality and lip-sync accuracy — is narrowing fast. Independent creators working with AI video generation tools are hitting similar trade-offs in audio sync, but at a fraction of the budget studio localization requires.

Pre-Production: Script Breakdown, Scheduling, and Storyboarding

Pre-production is where AI adoption is quietest but most consistent. A24 Labs, as part of the Google DeepMind partnership, is building storyboard generators that accept scripts as input and produce visual shot references. The system is designed to assist human directors and cinematographers rather than replace their creative decisions.

AI-assisted script breakdown — identifying every scene element that needs to be sourced, cast, or budgeted — is another area seeing adoption. Tools can parse a screenplay and auto-tag locations, props, stunts, and characters in minutes, a process that traditionally takes a first AD several days. Studios are cautious about sharing specifics, but breakdown automation is now common at the mid-tier production level.

The pattern here is consistent with what appears in the post-production examples: AI handles defined, bounded sub-tasks where output quality can be verified by a human expert. Storyboards get reviewed by directors; breakdowns get checked by first ADs. AI is generating a first pass, not a final decision.

The Union and Regulatory Layer

The WGA's 2023 contract established that if studios use WGA members' scripts to train AI models, they must compensate the writers. That's the first union agreement to treat AI training as a compensable act rather than a free input. It set the tone for subsequent negotiations across SAG-AFTRA and other guilds.

A Deloitte 2025 TMT report found that most studios remain cautious about generative AI for principal photography. The heaviest adoption is in pre-production (script breakdown, budget modeling) and post (dubbing, rotoscoping, sound). That matches what the production examples above actually show: AI is doing targeted, defined tasks, not replacing directors or cinematographers.

Frequently Asked Questions

What are the most well-known examples of AI in movies?

The clearest documented examples of AI in movies include ILM's de-aging work on Indiana Jones and the Dial of Destiny, Metaphysic's on-set deepfake work in Here, Respeecher's AI voice cloning for Darth Vader in Obi-Wan Kenobi, and Runway's rotoscoping on Everything Everywhere All at Once. Each used AI for a specific, bounded task rather than wholesale production.

Do studios need actor consent to use AI on their likeness?

Yes, in practice — and increasingly by contract. The SAG-AFTRA union has pushed for formal consent requirements, and the James Earl Jones/Lucasfilm arrangement (where Jones reviewed and signed over future rights) is the template studios cite. The proposed NO FAKES Act would make consent a legal requirement in the US.

Are major studios investing directly in AI companies?

As of mid-2026, yes. Studios have moved from simply licensing AI tools to taking ownership positions in them — Lionsgate’s partnership with Runway is one widely reported example. These are strategic stakes rather than vendor contracts.

How does AI in Hollywood compare to AI filmmaking tools for independent creators?

Hollywood's AI use is mostly narrow — de-aging one actor, rotoscoping one sequence, cloning one voice. Independent AI filmmaking platforms like Leyline automate the entire production pipeline: script, breakdown, asset creation, keyframes, and video generation with models like Veo 3, Seedance, and Kling. The tooling for small teams has in some ways outpaced what large studios have integrated end-to-end.

Keep Learning

For a broader view of how AI is changing production workflows, what is AI filmmaking covers the fundamentals. If you want to understand how the video generation models mentioned above compare for actual production use, Seedance vs Veo 3 breaks down the tradeoffs. For the emerging market these tools are powering, the vertical drama market explained gives context on where independent AI-produced content is heading commercially. And for the ongoing Hollywood debate, AI in Hollywood tracks the union and regulatory developments.