AI and film craft: new tools, older questions

Every generation of filmmakers has been handed a tool that older filmmakers found alarming — sync sound, colour, the handheld camera, digital editing, CGI. Each time, the same fear arrived: that the new tool would replace the craft. Each time, the truth turned out to be quieter. The tool changed what was easy, and the craft moved to wherever the hard, human part still lived. AI in filmmaking is the newest instance of a very old story, and we approach it the same way we would approach any powerful new lens: usefully, and with our eyes open.

What tools are good at

New generative tools are extraordinary at iteration. They let a small studio explore a hundred versions of a world in the time it used to take to draw one — testing a palette, a silhouette, a mood — so that by the time human hands commit, the choices are better considered. Used this way, in early visual development and exploration, they do not replace an artist; they give an artist more shots on goal. For an independent studio with more taste than budget, that is a genuine gift.

The tool changes what is easy. The craft moves to wherever the hard, human part still lives.

What should never be automated

But there is a line, and it is worth naming plainly. The parts of a film that carry meaning are the parts that must remain a human decision: what the story is about, whose life it asks you to inhabit, which moment the camera refuses to leave, the specific ache of a character's contradiction. These are not efficiency problems. A machine can generate a thousand faces; it cannot decide which one deserves your two hours, or why. Judgment, point of view, and responsibility do not automate, because they are precisely the things an audience is responding to when a film moves them.

The value of the made thing

There is also a reason we still build sets from felt and paper and move them one frame at a time. In an age when images are infinite and effortless, the made thing becomes more valuable, not less. Audiences can feel the difference between what was generated and what was cared for, and that felt difference is itself part of the story we are telling. Scarcity of effort is a language. When everything can be produced instantly, choosing to make something slowly, by hand, says something no shortcut can.

Older questions, held steady

So we are neither romantics who refuse the tools nor futurists who believe they change everything. We use what helps and refuse what hollows the work out, and the test is always the same question filmmakers have asked for a century: does this serve the story, or does it just save time? Time saved is only worth it if the human parts — the judgment, the point of view, the care — are protected and poured back into the film. New tools, older questions. The questions are the craft.

For what those older questions look like in practice, see what is visual development in film.

Frequently asked questions

How is AI used in filmmaking?
Mainly for rapid iteration in early visual development — exploring many versions of a world, palette or design quickly — so human artists can make better-considered final choices. It is a tool for exploration, not a replacement for judgment.

What parts of filmmaking should not be automated?
The meaning-bearing decisions: what a story is about, whose life it asks you to inhabit, and the specific emotional choices within a scene. Judgment, point of view and responsibility do not automate.

Does AI-generated imagery devalue handmade film craft?
If anything, the opposite. When images become infinite and effortless, the visibly made thing gains value, because audiences feel the difference between what was generated and what was cared for.