October 1, 2026
For the past couple of years, much of the AI filmmaking conversation has revolved around capability.
Can it make convincing video? Can it maintain a character? Can it create a usable shot? Can one person make something that once required a crew?
Those questions haven’t disappeared.
But this week, something else stands out.
AI is moving out of the demo reel and into the workflow.
Runway is integrating directly with DaVinci Resolve. Film schools are teaching AI-assisted production. Filmmakers are discovering what AI production actually costs once a project has clients, revisions, and professional delivery requirements.
Meanwhile, established producers are experimenting with generative production, governments are investing in AI-video industries, and thousands of AI films are showing up at festivals.
The technology is beginning to collide with actual filmmaking.
That’s where things get interesting.
IN THIS ISSUE
Runway moves into Resolve
Generative video gets closer to the timeline.
Film schools start teaching the new workflow
AI literacy is becoming part of production education.
The $80,000 AI film
Cheap generation doesn’t necessarily mean cheap professional production.
QUICK BRIEFS
The Call of the Wild tests generative production • China’s AI-video boom meets a reality check • 8,067 films enter an AI festival • CineMe tries connecting development to production.
RUNWAY MOVES INTO DAVINCI RESOLVE
FACT
Runway released a DaVinci Resolve plugin on September 23, extending the company’s existing integrations with Premiere Pro and After Effects.
The Resolve integration lets filmmakers work from footage already under the playhead, generate or modify material using Runway tools, including Aleph 2.0, compare the result, and return generated footage directly to the Media Pool and timeline.
The plugin is available for macOS and Windows and requires DaVinci Resolve Studio 19 or later. Generation requires a paid Runway plan and credits.
AFFECTS
EDITORS • DIRECTORS • POST
The important development isn’t that Runway has invented another way to generate video.
It’s where they’re putting it.
The less filmmakers have to export footage, open another application, upload material, generate something, download it, and then bring everything back into the edit, the more useful these tools become in an actual post workflow.
That’s a considerably different proposition from playing with a video generator in a browser.
CONSEQUENCE: Generative video is getting closer to the place where filmmakers actually make decisions: the timeline.
SOURCE: Runway — Using the Runway Plugin in DaVinci Resolve
FILM SCHOOLS START TEACHING THE NEW WORKFLOW
FACT
AI-assisted filmmaking is increasingly showing up inside film and media education.
Runway CEO Cristóbal Valenzuela told Semafor that studios have been asking where they can find people who already understand AI filmmaking tools rather than training existing teams from scratch.
That characterization comes from the CEO of a company selling those tools, so it deserves that attribution.
The education trend itself, however, is visible.
Runway-related filmmaking courses have appeared at institutions including USC, RISD, MIT, Harvard, and NYU. Google has also been running filmmaker programs around Flow.
Other programs are moving beyond simple prompting. UCLA Extension coursework, for example, includes image generation, performance capture, compositing, upscaling, audio and other parts of an AI-assisted production workflow.
AFFECTS
STUDENTS • EDUCATORS • EMERGING FILMMAKERS
The question for film schools is rapidly becoming less:
Should we teach AI?
And more:
What the hell should we teach about it?
Knowing how to prompt a model isn’t filmmaking.
Students still need composition, editing, sound, continuity, performance, collaboration, ethics, storytelling, and judgment.
AI literacy may simply become another part of that education.
CONSEQUENCE: Students entering production may increasingly be expected to understand AI-assisted workflows without confusing knowledge of the tools with knowledge of the craft.
SOURCE: Semafor — Hollywood scouts film schools for AI-native talent
THE $80,000 AI FILM
FACT
Robert Gaudette created the seven-minute, 50-second AI short A Face Only a Mother Could Love largely on his own over roughly two weeks.
His workflow included tools such as Grok, Nano Banana, and Seedance.
The film went on to win the Grand Prix at Runway’s AI Film Festival, the Reply AI Film Festival in Venice, and other recognition.
That sounds like another story about AI dramatically lowering filmmaking costs.
Except Gaudette offered an important reality check.
He estimated that producing an equivalent eight-minute project for a client could cost around $20,000.
Bring that same work to a broadcast-level finish—with revisions, frame repairs, legal requirements, and professional delivery—and his estimate rises to approximately $60,000–$80,000.
He also acknowledged something that rarely makes it into the AI sales pitch:
Sometimes it's simply cheaper to shoot the damn thing.
REALITY CHECK
DIRECTORS • PRODUCERS • AI FILMMAKERS
Making something for yourself and delivering something professionally for somebody paying you are very different propositions.
Professional filmmaking includes revisions, approvals, consistency, rights, finishing and delivery.
AI doesn’t magically erase those things.
And neither does it eliminate judgment.
A model can generate fifty possibilities.
Somebody still has to know which one belongs in the movie.
CONSEQUENCE: AI can lower certain production barriers without automatically making professional filmmaking cheap.
SOURCE: Forbes-syndicated reporting / Runway / Reply AI Film Festival
QUICK BRIEFS
THE CALL OF THE WILD BECOMES AN AI PRODUCTION EXPERIMENT
MIGHT AFFECT • PRODUCERS • DIRECTORS • VFX
Producer David Barron, whose credits include six Harry Potter films, has joined Simon Rumley on a new adaptation of Jack London’s The Call of the Wild.
The project is being developed at nmatic and plans to combine generated imagery with CGI and traditional VFX.
The filmmakers say AI may help them approach the story through Buck’s subjective experience while tackling environments and animal action that would otherwise be difficult and expensive.
That’s their production argument. We don’t yet know how much of the finished movie will be generated versus conventionally produced, and there’s no finished film to judge.
CONSEQUENCE: An established producer is testing generative tools against a specific storytelling and production problem, not simply making an AI demonstration.
SOURCE: Deadline
CHINA’S AI-VIDEO BOOM GETS A REALITY CHECK
REALITY CHECK • INDUSTRY • PRODUCERS
Chinese cities are competing to attract AI-video creators and companies with subsidies, computing resources and other incentives as the country tries to develop AI-generated entertainment into a larger industry.
But Reuters reported another number that’s considerably more interesting.
About 221,900 new AI shows launched on Douyin during the first half of 2026.
Only 1,055 exceeded 100 million views.
That’s roughly half of one percent.
CONSEQUENCE: AI may dramatically increase how much content can be produced. It doesn’t increase the amount of attention available to watch it.
SOURCE: Reuters
8,067 AI FILMS WALK INTO A FESTIVAL
AFFECTS • FILMMAKERS • FESTIVALS
The Astana AI Film Festival, running October 1–3 in Kazakhstan, received 8,067 film submissions from 125 countries and regions.
Organizer materials identify approximately 4,756 individual creators, meaning some filmmakers submitted multiple projects.
The festival is offering $1 million in prizes along with a separate $1 million production investment fund.
Forget whether every one of those films is good.
Eight thousand films showed up.
God help the screening committee.
CONSEQUENCE: AI filmmaking has developed enough participation to support its own international festival, competition and funding ecosystem.
SOURCE: Astana AI Film Festival
CINEME WANTS TO CONNECT THE SCRIPT TO THE BUSINESS
WATCH • PRODUCERS • DIRECTORS • DEVELOPMENT
CineMe is scheduled to launch today, October 1.
Founded by filmmaker Dan Hartley and former Prime Video UK managing director Chris Bird, the platform aims to connect script visualization, storyboards, previs and pitch materials with less glamorous—but rather important—things such as budgeting, scheduling and financing.
Former Sky film executive Julia Stuart has joined its advisory board.
That’s potentially more interesting than another standalone image generator.
But today’s launch means the claims now need to meet the product.
We’ll watch it.
CONSEQUENCE: If CineMe delivers what it promises, AI-assisted filmmaking may begin connecting creative development with production planning rather than stopping at generated images.
SOURCE: Advanced Television
WHAT IT MEANS FOR FILMMAKERS
A common thread runs through most of this week’s stories.
Integration.
Not another model generating a slightly prettier image.
Runway wants to live in the edit.
Film schools are beginning to incorporate AI into production education.
AI filmmakers are discovering the difference between making something and delivering it professionally.
Feature producers are testing the technology against specific production problems.
And new platforms are trying to connect creative development with budgeting, scheduling and financing.
That’s a healthier stage for this technology.
Because once AI enters an actual filmmaking workflow, it has to survive something far less forgiving than a demo:
Filmmakers.
WORTH YOUR TIME
If you’re working in post, the Runway/DaVinci Resolve integration is worth looking at—not because you need to use it tomorrow, but because it shows where generative tools are heading.
Pay attention less to what Aleph can generate and more to how many steps disappear between the generated material and your timeline.
That’s the workflow story.
IGNORE THIS WEEK
Any headline claiming AI has made filmmaking essentially free.
Robert Gaudette’s experience reminds us that generating images is only one part of production.
Clients still change their minds.
Shots still break.
Deliverables still have specifications.
Lawyers still exist.
Sorry.
THE CINESIGHT TAKE
We’ve spent the early years of generative filmmaking asking what AI can make.
That’s understandable.
The first time you type a sentence into a computer and watch it produce moving images, it’s difficult not to stare at the trick.
But the trick is becoming less important.
The more interesting question now is:
Where does this actually belong in filmmaking?
Sometimes the answer may be previs.
Sometimes post.
Sometimes concept development.
Sometimes a shot that would otherwise be prohibitively expensive.
And sometimes the answer will be:
Nowhere. Shoot the damn thing.
That’s not anti-AI.
That’s filmmaking.
Tools earn their place because they solve problems—not because they’re new.
And as AI moves out of the demo reel and into real production workflows, we’re finally getting enough evidence to start figuring out where it actually belongs.
ABOUT THIS REPORTING
CineSight uses AI as part of the research and production process, but significant factual claims are checked against available reporting and primary sources before publication.
When something is confirmed, we say so.
When it’s somebody’s claim, we attribute it.
When we don’t know yet, we’d rather say we don’t know yet than manufacture certainty.
Seems like a pretty good policy for filmmaking too.
ABOUT CINESIGHT BRIEFS
CineSight Briefs follows the technology, tools and changes affecting working and emerging filmmakers.
No AI cheerleading.
No AI apocalypse.
No pretending every new model has “changed filmmaking forever.”
Just:
What happened.
Who it affects.
And whether it’s worth giving a shit about.
See you in the next Brief.
— Rob Tuscani
CineSight