Why are film studios asking graduates if they already know AI video tools?

Studios are now scouting film schools for graduates who already know how to use AI video tools instead of training their own staff. Runway CEO Cristobal Valenzuela told Semafor that studios keep asking how to find people who already know how to use the software, ending a century of apprenticeship where the industry taught newcomers on the job.

What does AI fluency really mean in the film industry?

AI fluency usually certifies skill with a specific vendor's software, which is a one-week skill, rather than the ten-year craft of knowing what a shot should look like. Studios like Netflix have asked every employee to build AI fluency, but the software changes quarterly while the eye for good work does not.

Who pays for AI filmmaking training: the studio, the vendor, or the student?

The student pays tuition, the vendor writes the course, and the studio hires the graduate so it does not have to train its own staff. Runway launched its first university course at NYU Tisch and powers USC's graduate AI filmmaking course, so the vendor collects the tuition pipeline now and the enterprise contract later.

The prerequisite -- CinePrompt Field Notes

For a hundred years, a person who wanted to work in film showed up knowing almost nothing and expected to be taught. The studio, the union, the old-timer at the next bench. Somebody showed you where the light went and why, and the teaching was understood to be part of the job. That arrangement ended quietly this week, inside a single question reported by Semafor.

Studios are now scouting film schools for graduates who already know how to use AI video tools, rather than training their own staff. "How do we find people who already know how to use this?" Runway chief executive Cristóbal Valenzuela said studios are asking. The question carries a second, unstated half. We do not want to pay to teach you.

A studio is an organization that has always trained people. It trained its editors, its colorists, its assistant directors, its story artists. Apprenticeship was a cost the industry carried because the skill could not be bought off a shelf. The moment the skill became software, the studio reclassified it. It is now a prerequisite rather than a responsibility. You arrive with it, or you do not arrive.

The prerequisite has a curriculum, and the curriculum has an author. Runway announced its first university course at NYU Tisch in December 2024, inside the Martin Scorsese Virtual Production Center. USC's School of Cinematic Arts runs a graduate AI filmmaking course built on Runway's tools. The software is taught at RISD, at MIT, at Harvard, and at universities across Europe and Latin America. Google runs Flow Sessions. OpenAI funds short films through Tribeca. The vendor writes the course, the student pays the tuition, and the graduate walks out fluent in the vendor's software.

The arithmetic is not subtle. The tool company seeds the schools so that graduates are fluent in its product. The studio hires those graduates so that it does not have to train its own staff. The student sits in the middle, paying to be taught, and later serving as the thing the studio rented rather than the person it invested in. The vendor wins twice. It collects the tuition pipeline now and the enterprise contract later.

Then there is the question nobody at the scouting table is asking. What does "AI-native" actually certify? Fluency with the software is a one-week skill. Knowing what the shot should look like is a ten-year one. The software changes quarterly. The eye does not. A studio that hires for Runway fluency is hiring the half that goes stale and treating the half that lasts as the student's private business.

Brad Tangonan, who went through Flow Sessions, put it plainly: "You're going to have to learn how to use these tools if you do want to be in the industry." He is right, and he is also describing the trap. The individual hears the prerequisite and complies, because the individual needs the job. The studio hears the prerequisite and is relieved, because the compliance costs nothing. The vendor hears the prerequisite and says nothing, because it wrote the exam.

The same week, Google made 1080p AI video generation free to anyone with a Google account, inside Google Vids. The tool is now nearly free to touch. That is what makes fluency a prerequisite rather than a purchase. Nobody charges you for the skill anymore. They just expect you to have it.

Netflix has used generative AI in roughly three hundred titles and has asked every employee to build "AI fluency." The same logic, delivered as a memo. Fluency is now a standing order for the people already in the building and a hiring criterion for the ones not yet through the door. Nobody is paying for the class. The class is assumed.

Every technology transition has a moment like this one. The new tool arrives. The schools teach it. The studios hire for it. For a little while everyone mistakes the tool for the craft. Then the tool changes, and the people hired for the tool get retrained, and the people hired for the craft keep working. The cycle is old. What has changed is who pays for it.

The film school degree is quietly becoming a proxy for "already fluent in Runway." The job market is pricing the navigation, not the vocabulary. The vocabulary still cannot be bought, and now it is not even being looked for.

Bring it anyway.

Bruce Belafonte is an AI filmmaker at Light Owl. He learned the buttons and then watched them move.