Film Production AI in 2026: How Hollywood and Indie Filmmakers Are Actually Using It
In 2024, most conversations about film production AI were hypothetical. People argued about whether a machine could ever "make a movie." In 2026 the argument has moved on. Martin Scorsese storyboards with generative AI. Netflix has shipped AI-built VFX in a hit series. A 95-minute feature made entirely with AI screened at an industry event during Cannes. And one of the most respected directors working today was open about using AI in his latest project, much to the dismay of the online community.
The real question for filmmakers is no longer whether AI belongs in film production. It's where in the process it actually helps, and where it backfires. This guide covers the most notable AI film projects of the past year, what directors and creators are saying about them, and a practical way for indie filmmakers to use AI without losing their voice or their crew's trust.
What Is Film Production AI?
Film production AI is the set of machine-learning tools used across the filmmaking pipeline: script analysis, shot listing, storyboarding and previs in pre-production; on-set tools like smart camera tracking; and VFX, editing, dubbing and localization in post. Some tools generate images and video. Others do the slow, structural work, like breaking a screenplay into scenes and shots, so filmmakers can spend more of their time on creative decisions.
Here's how it maps to the stages of a production:
Development: script coverage, a first pass on legal review, and proof-of-concept trailers for pitching.
Pre-production: AI shot lists generated from a screenplay, storyboard frames, previs, mood and reference images, location and prop concepts.
Production: autofocus and subject tracking, virtual production backgrounds, on-set reference, drone shots.
Post-production: VFX shots, cleanup, color assistance, transcription and editing aids.
Distribution: AI dubbing, lip-sync and localization for smaller territories.
Six High-Profile Examples of AI in Film Production (2025–2026)
1. Martin Scorsese Storyboards with Generative AI
In June 2026, Martin Scorsese said he uses FLUX, a generative image model from Black Forest Labs (which he advises), for storyboarding and previs. He isn't using it to put AI images on screen. He's using it to communicate.
"Now, with this tool, I can share what I'm visualizing more clearly and efficiently to my creative team." — Martin Scorsese
He also said AI previs can mean "less wear and tear on the crew," and added: "Cinema is a young medium, only around 125 years old, so we have to be open to how it can evolve."
2. Netflix's The Eternaut Used AI for a Key VFX Shot
Netflix's Argentine sci-fi series The Eternaut (El Eternauta) used generative AI to create a building-collapse sequence in Buenos Aires. Co-CEO Ted Sarandos told analysts the shot was completed "ten times faster than it could have been completed with traditional VFX tools and workflows." His framing: AI is "real people doing real work with better tools," and an opportunity to make films "better, not just cheaper."
3. Hell Grind: A Fully AI-Generated Feature, Made in Two Weeks
Higgsfield AI backed Hell Grind, an action feature directed by Aitore Zholdaskali. A 15-person team generated every frame for about $500,000, with roughly 80% of the budget going to compute. It screened at an industry event in Cannes in May 2026 (not as part of the official festival).
The interesting part is how it was made. The team generated 15-second clips and regenerated them until they matched. Prompts averaged about 3,000 words per shot to keep characters and locations consistent. Higgsfield's CEO compared parts of the process to "a slot machine." Co-founder Mahi De Silva told Fortune: "What we have found is that filmmaking is still an art."
The takeaway for filmmakers: even a fully AI-generated movie lives or dies on shot planning. A 3,000-word prompt is a shot description. Whoever can define the shot clearly (subject, framing, movement, continuity) gets usable footage. Everyone else just pulls the lever again.
4. Runway and Luma Move into Series and Shorts
AI video companies are now making content themselves. Runway made the short Rules of Horse… for Dummies in under two weeks and published its character sheets and prompts so others could build on them. Luma partnered on The Old Stories: Moses, a hybrid live-action/AI series starring Ben Kingsley and backed by AWS.
"These are tools that can amplify unique points of view." — Jamie Umpherson, Chief Creative Officer, Runway
"The job will still be done by humans, but AI will be doing parts of the task." — Amit Jain, CEO, Luma
5. Amazon MGM Studios Is Building AI into Pitching
At SXSW 2026, Amazon MGM Studios executives described an internal tool called Pitch Lab that helps creators build proof-of-concept trailers instead of pitching on paper. Dan Scharf, Head of Global Business Operations, said AI in VFX and production design "allows you to get more scope than the budget will allow otherwise," and that creators need to be "a tester and a product experimenter to be successful."
6. Darren Aronofsky's On This Day… 1776, and the Backlash
Not every experiment landed. In January 2026, Aronofsky's AI studio Primordial Soup released On This Day… 1776, a short-form Revolutionary War series made with Google DeepMind and TIME Studios. Professional voice actors did the narration, but the AI-generated visuals drew harsh criticism. Critics called it "AI slop" and pointed to uncanny-valley faces.
The lesson: audiences are far more forgiving of AI they never see. Using AI to help plan, previs or speed up a VFX shot rarely starts a fight. Putting obviously synthetic images on screen often does.
What Filmmakers and Creators Are Saying About AI in Film
Among working filmmakers, the attitude has shifted from fear to cautious, practical curiosity:
"I don't think [using] AI is cheating. As long as you have something to say, I don't care what you use." — Roger Deakins, cinematographer (source)
AI is "something to embrace, not fear," with the goal of "integrating these technologies responsibly and with still the primacy of the creative voice." — Shawn Levy, director of Star Wars: Starfighter (source)
We've been hosting these conversations since 2024. At our NAB New York panel on generative AI and the cinematographer, cinematographer David Kruta, founder of the 6,000-member Cinematography Salon, described using AI to design a shape-shifting alien creature in pre-production:
"I don't have to rely solely on drawings or abstract ideas. With AI, I can create iterations of a specific object in minutes, show them to the director, and we can decide what best fits."
Moderator Gary Adcock put it simply: "This is building ways to make our productions more efficient, our knowledge more useful, and be able to deliver on a timely fashion."
At our AI Reshaping Media panel during NY Tech Week, speakers from Luma AI, B&H and Pepper Content raised the same tension the industry is still working through. AI can give independent creators Hollywood-level tools at a fraction of the cost, but questions about training data, consent and craftsmanship are real and haven't been settled.
The Pattern: AI Works Best in Pre-Production
Look across these examples and a clear pattern shows up. The projects that drew praise used AI behind the camera: Scorsese's storyboards, Amazon's pitch trailers, Netflix's single VFX shot. The projects that drew backlash put AI in front of the audience.
That makes pre-production the place where AI gives the most help for the least risk:
It's invisible to audiences. Nobody watches your shot list.
It saves the scarcest resource. On indie shoots, prep time is always shorter than it should be. A locked shot list is a luxury.
It improves communication. As Scorsese said, the value is sharing the picture in your head with your DP, AD and crew faster.
It keeps humans in charge. A first draft you edit is still your vision. That matters creatively, and as Amazon MGM's legal head pointed out at SXSW, human involvement also matters for copyright protection.
How Indie Filmmakers Can Use AI in Pre-Production Today
You don't need a studio budget or a 3,000-word prompt to get the benefit. Here's a practical workflow with ShotKraft, the AI shot-listing tool we built for working filmmakers:
Upload your script. ShotKraft reads your screenplay and generates a first-draft shot list for every scene in minutes. Building one by hand for a feature usually takes days (here's how to do it by hand if you want to compare).
Edit like a director. Treat the AI draft as coverage suggestions. Change shot types, framing, camera movement, lenses and notes. The creative calls stay yours.
No script? Shotlist anything. Paste a client brief, a music video concept, a short story or a news item, and ShotKraft writes a visual treatment and shot-lists it. See how it works.
Add visual references. Generate reference images per shot so your DP and production designer can see the plan, the same idea as Scorsese's AI storyboards, built into your shot list. (Shot list vs storyboard: which do you need first?)
Share it with your crew. Export a clean PDF or CSV, or send a link, so everyone shows up on set working from the same plan.
If you're comparing tools, our guide on how to choose a shot list maker covers what to look for.
Guardrails for Using Film Production AI Responsibly
Keep a human in the loop. Use AI for drafts and references, then make the decisions yourself. It's better creatively and stronger legally.
Be careful with likeness and voice. Only use a real person's likeness or voice with clear, written consent.
Check your vendors. Lawyers at SXSW 2026 warned that smaller AI providers may not be able to cover liability or offer indemnification. Read the terms before AI output goes into a deliverable.
Be thoughtful about what goes on screen. The 1776 backlash shows that audiences notice. Use AI where it strengthens the work, not where it replaces the craft people came to see.
Frequently Asked Questions About Film Production AI
How is AI used in film production?
AI is used across the pipeline: script analysis and shot listing in pre-production, storyboards and previs, smart tracking on set, VFX and editing in post, and dubbing and localization in distribution. The most widely accepted uses today are in pre-production and VFX, where AI speeds up the work without replacing the people doing it.
Will AI replace filmmakers?
Nothing in 2026 suggests it will. Even the fully AI-generated Hell Grind took a 15-person team and extremely detailed shot descriptions. Directors including Scorsese, Deakins and Levy describe AI as a tool that supports the creative voice, not a replacement for it.
What's the best way for an indie filmmaker to start using AI?
Start in pre-production. Generate a shot list from your script, refine it, add visual references and share it with your crew. It saves days of prep, and it's invisible to your audience.
Is AI-generated footage allowed in festivals and on streaming platforms?
Policies differ by festival and platform and are still changing, so check each one's current rules before submitting. AI used only in planning (shot lists, storyboards, previs) doesn't appear in the final picture and is generally not an issue.
Plan Your Next Shoot with AI, Your Way
AI isn't going to make your film for you, and the best filmmakers don't want it to. What it can do is give you back the prep time you never have enough of, so you show up on set with a clear plan and a crew that understands it.
Turn your script into a shot list in minutes. No credit card needed to start.
Comments