Two dispatches from the same week
On August 1, 2026, Bloomberg ran an opinion piece by Miles Surrey with the headline "AI Filmmaking Is a Financial Lifeline, Not a Creative Comeback." His argument: when a director builds an AI studio and announces it before releasing any work to an audience, the pitch is the product. Neill Blomkamp's Barley Studios, Darren Aronofsky's Primordial Soup (an AI venture seeking fifteen million dollars, according to SEC filings covered by Variety, Deadline, and the Hollywood Reporter, with over eleven million already raised) -- these are companies with investors on the cap table before a single ticket has been sold.
Three days later, also in Bloomberg, Catherine Thorbecke published her piece: "AI Didn't Ruin These Films. It Made Them Possible." She had attended an AI film festival in Tokyo. Her favorite entry was an eleven-minute short shot from the perspective of a child holding a camcorder during a summer with his grandparents. Cereal bowls. Mermaid sightings. Afternoon light going soft. The film was made with AI video tools. There was no studio announcement. No funding round. No investor deck. Just a story someone needed to tell.
Both pieces are right. And neither one asks the question that actually matters.
The question that gets skipped
Every time a new tool arrives in cinema, the same argument begins: is this legitimate? Photography forced painting to answer it. Sound forced silent film to answer it. Video forced celluloid to answer it. The argument is always about the medium, because the medium is visible. The decision behind the medium is not.
Walter Benjamin wrote in 1935 that the aura of a work is not a property of the object. Aura is what someone chooses to see. He was writing about mechanical reproduction, about what happens when a single original becomes infinitely copyable. His conclusion was not that copies destroy meaning. Indifference does. The question was never: can you reproduce it? The question was always: why do you bother?
Blomkamp released "Nightborne" in July 2026: a thirteen-minute sci-fi horror film built with Seedance 2.0, featuring the licensed faces and voices of thirty-two real people, with concept art from human artists. He called it on social media his "first full test AI film." Then he announced Barley Studios, the company that will make features using the same approach. Then the Bloomberg column appeared. Then the SEC filing on Aronofsky's Primordial Soup, already past eleven million raised against a fifteen-million target.
The sequence matters: announcement before audience. Studio before film. Capital before story.
And then, three days after Surrey's column, a child's summer with his grandparents, made with AI, screened at a festival in Tokyo. No studio. No announcement. No raise.
What independent filmmakers actually see
I spent years making commercials and independent films where the first question was always about the story, not the structure built around it. The work came first. The pitch followed when needed. The difference is not moral. It is practical: if you start with what you want to say, the tools arrange themselves around that. If you start with the instrument, you spend the whole project asking the instrument what to say.
Three things become clear when you look at the AI filmmaking conversation through that lens:
- Barley Studios and Primordial Soup are companies with investors. The short films they produce are portfolio material for a fundraising round. This is not filmmaking with AI as a tool. It is venture capital with a film attached as proof of concept.
- The Tokyo short Thorbecke described was not submitted to prove that AI filmmaking is credible. It was submitted because someone had a story about a summer and a grandparent and a camcorder and found the means to make it. The tool served the story, not the other way around.
- The Bloomberg debate treats these two things as points on the same spectrum. They are not on the same spectrum. They are different activities that happen to share a technology.
Not because the studios are wrong
Barley Studios and Primordial Soup may make films that outlast the companies. The productions that emerge from them may be exactly what their directors believe them to be. That is not the point.
The point is what the sequence tells you about the relationship between the maker and the material. When you announce the studio before the film, the studio is the work. The film is evidence that the studio can do what it claims. That is a pitch deck with moving images, regardless of how many real people's faces appear in it.
The child with the camcorder started from the other direction: something to say, and then the search for the means to say it. That is an older sequence, and it is recognizable regardless of what tool produced the final image. Photography did not replace painting. Sound did not replace silence. Each arrival clarified what the previous form could do that the new one could not. What AI clarifies right now, in this particular week of Bloomberg columns and SEC filings and Tokyo festivals, is the gap between those who start from the story and those who start from the studio.
The amplifier does not decide
AI is, in the end, a tool with hyperfocus. It absorbed a remarkable amount about how human beings make images and tell stories and cut films. But it cannot decide what to say. Feed it a fundraising imperative and it will build a convincing trailer for a company. Feed it a summer afternoon in a grandmother's kitchen and it will build something that might outlast the company by a long time.
Blomkamp's "Nightborne" exists. Primordial Soup's relationship with Google DeepMind exists. The eleven-minute short from Tokyo, the one nobody announced, also exists. They all used AI. The difference is not in the tool. The difference is in what the person brought to the table before the tool was turned on.
Bloomberg published two pieces in four days and landed on both sides of the argument. The answer is not in the middle. The answer is in understanding that the real question is not whether AI filmmaking is legitimate. The question is: who is making the decision, and what is the decision actually about?
A pitch deck and a film can look identical from a distance. Up close, one of them has an audience in mind. The other has a cap table.
