How AI Studios Are Rewriting the Rules of Movie Production

Rows of computers replaced the usual bustle of a film set when actor Tori Thomas waited for her call during early shooting for Touch Grass, an AI-enabled horror film. The production comes from Promise, a new studio building movies with AI models that create backgrounds, special effects and even “synthetic performers.”
Promise is backed by Google, Silicon Valley venture capitalists and Disney, placing it among a rising number of AI film studios taking root in Hollywood. Its work reflects a split taking shape across the industry: some filmmakers see AI as a powerful production tool, while others fear what it may do to creative jobs and the people who depend on them.
Promise bets on faster, cheaper filmmaking
Touch Grass has a budget in the low millions of dollars, and its production uses Seedance 2.5 to generate backgrounds and effects. The model, released this month, can churn out about 30 seconds of video clips, giving filmmakers a way to build parts of a movie without creating every element through traditional production methods.
That approach helps explain why Promise has attracted backing from major technology and entertainment names. The studio is feeding off the potential of rapidly accelerating AI models, using them across the filmmaking process instead of limiting them to one visual effect or editing task.
Joel Hynek, who supports the technology, compared the change with the move from optical film to digital filmmaking. “When we went from optical to digital film, we thought we’d gone to heaven. It’s like that again. So many things are easy to do. I am an embracer of AI. Some people are wary of it, a little afraid of it, but I see the incredible promise.”
That promise comes with a direct warning about jobs. Laura Friedman said, “We can’t wait until we see tens of thousands of American workers displaced.” Her comment captures the concern surrounding tools that can create performances, locations and effects while reducing the need for parts of a conventional shoot.
AI moves from experiments into major releases
The technology is no longer limited to small demonstrations. Netflix said it had used AI in 300 of its 1,000 titles in 2026, showing how deeply the tools have entered mainstream production.
Ron Howard is set to release an AI-enabled animated documentary about prisoners of war in Vietnam, working with AI studio Obsidian. The project adds a documentary to the growing list of formats testing AI, alongside horror films, animated work and online showcases.
Brad Pitt has also argued that AI can help movies that struggle to secure financing. “AI is going to be the very thing that, if used as a tool, is going to help these mid-budget films, $40m- to $90m-budgeted films, get made.”
The argument is not that AI will replace every part of filmmaking. Instead, supporters present it as a way to make projects possible when traditional budgets, studio decisions or production demands stand in the way. Kavan “the Kid” Cardoza predicted a creative renaissance that would circumvent studio giants, pointing toward a film world where more creators can develop projects outside the biggest companies.
A movie made from prompts and likeness agreements
Higgsfield AI has offered one of the clearest examples of this new production model with Cully Hill Boys, an AI movie created to showcase AI filmmaking tools and promote its Seedance 2.5 functionality. Higgsfield uploaded the film online on Aug 13, 2026, after the project grew from a script initially written by Tim Planagan.
The film follows three struggling rappers from East London who steal a crime lord’s money. Adilet Abish and Aitore Zholdaskali co-directed the movie, which stars Mikyle “N3on” Rafiq, Matt Kiatipis and Israel Adesanya.
None of those actors shot scenes on set. Instead, they signed licensing agreements for the use of their likenesses, allowing the film to include their recognizable identities within an AI-generated production.
That detail separates Cully Hill Boys from a conventional movie and puts the question of performance at the center of the debate. The film combines licensed likenesses with generated scenes, backgrounds and effects, showing how an AI project can use familiar performers without asking them to take part in a traditional shoot.
The prompts used to create the film also reveal how filmmakers are guiding AI systems. They included references to Guy Ritchie and Edgar Wright, along with terms such as “GUY RITCHIE COVERAGE” and “DYNAMIC clip-style (Edgar Wright snap).” Those instructions aimed to shape the way scenes were framed and edited while the production relied on AI tools.
Promise’s Touch Grass and Higgsfield’s Cully Hill Boys point to two sides of the same shift. One uses AI during an early film shoot with a human actor waiting for her call; the other builds a movie around licensed likenesses and generated footage.
The industry reaction remains mixed, but the direction is clear. AI studios are developing projects, established companies are using the tools, and filmmakers are testing what happens when backgrounds, effects and performances can begin with a prompt.
Based on




