Let’s clear up the biggest misconception first.
Using AI in the creation of a film does not automatically make the finished film uncopyrightable. But what the human creator actually contributes matters enormously.
People are confused. There is an astonishing amount of bad information floating around. And we have something most companies don’t: a real operational process for documenting human contribution.
That sentence alone would save the internet several million arguments.
The real issue is much more specific:
What did the human creator actually create?
That is the question sitting underneath the U.S. Copyright Office’s current approach to generative AI.
The Copyright Office has made clear that copyright protection remains grounded in human authorship. At the same time, it has also made clear that using AI as an assistive tool, or including AI-generated material inside a larger human-authored work, does not automatically prevent copyright protection.
That distinction matters enormously for filmmakers, writers and visual creators.
AI-ASSISTED IS NOT THE SAME AS AI-GENERATED
Imagine two different filmmaking processes.
In the first, someone types a prompt into an AI image generator, accepts the first result and publishes it exactly as generated.
In the second, a filmmaker develops the story, establishes the visual language, generates dozens of possibilities, rejects most of them, modifies selected material, composites elements, edits the images into sequences, determines pacing, adds sound, creates typography, makes narrative choices and assembles everything into a finished film. Those are not the same creative process.
Copyright law cares about that difference.
The Copyright Office’s January 2025 report says copyright may protect human-created expression in works containing AI-generated material, including creative human modifications and arrangements. It also says that prompts by themselves generally do not provide enough human control over the resulting expressive elements to make the AI output copyrightable. In normal-person English: The prompt is not the magic part. The human creative work is.
WHAT HUMAN AUTHORSHIP CAN LOOK LIKE IN AN AI-ASSISTED FILM
Filmmaking has always involved layers of authorship.
A director doesn’t personally manufacture the camera.
An editor doesn’t need to have created the footage.
A production designer doesn’t need to have grown the trees used to build the set.
Creative authorship often comes through selection, arrangement, transformation and expression. AI-assisted filmmaking raises new questions, but some of the underlying creative principles are surprisingly familiar.
Human contribution may include things such as:
- Writing the screenplay.
- Designing characters and visual concepts.
- Selecting among generated alternatives.
- Rejecting unusable material.
- Manually altering images.
- Combining or compositing visual elements.
- Determining shot order.
- Editing scenes.
- Establishing timing and pacing.
- Choosing music and sound.
- Creating sound design.
- Designing titles and typography.
- Determining the final narrative arrangement.
- Making substantive revisions.
The key is not simply that a human touched the project.
The human contribution needs to contain actual creative expression.
THE PROMPT PROBLEM
This is where some creators get a nasty surprise.
A prompt can be extremely detailed.
You can describe the lighting, lens, wardrobe, camera angle, mood, texture, composition and
expression.
That may require skill.
But under the Copyright Office’s current analysis, prompts generally function more like instructions to the system than direct human control over the expressive elements of the resulting output. That doesn’t make prompting worthless. Far from it.
It simply means creators should stop assuming that writing an excellent prompt automatically gives them copyright in everything the model generates. The much safer creative mindset is: Generate. Select. Transform. Arrange. Edit. Create.
That is where human authorship becomes much easier to identify
WHY GREENLIT DOCUMENTS THE HUMAN WORK
This question became important enough to us that we built it directly into the Greenlit Studio production process.
For client projects, we maintain what we call a Human Creative Contribution Record. It documents the human-directed work performed during production.
For example:
“9/18/26 | Reviewed 34 character generations; rejected 27; selected 7; manually modified facial details and wardrobe; established final character reference.”
That record might also document editing sessions, visual changes, compositing decisions, typography, sound design, story revisions or final assembly
Does keeping such a record magically guarantee copyright protection?
No.
Copyrightability depends upon the work itself and the applicable law.
But if the central legal question is human authorship, it makes enormous sense to know what the humans actually did.
Memory is a terrible production management system.
Documentation is better.
WHAT ABOUT REGISTERING THE WORK?
When a work submitted for registration contains more than a minimal amount of AI-generated material, the Copyright Office may require disclosure of that material and identification of the human-authored portions being claimed.
This is an area where creators should read the Copyright Office’s current guidance carefully and, for valuable intellectual property or complicated situations, talk with a copyright attorney.
Greenlit Studio is not a law firm.
We are a filmmaking production studio.
Our job is to build our filmmaking process in a way that keeps human authorship visible, intentional and documented rather than treating the human as the person who merely typed “cinematic” into a prompt box.
THE REAL OPPORTUNITY
There is a larger point here that tends to get buried under the panic. AI does not make human creativity less important.
It may make it easier to produce raw material.
That actually makes creative judgment more valuable.
When anyone can generate 100 images, selecting the right one matters.
When anyone can create a video clip, knowing where it belongs matters.
When anyone can produce something beautiful, knowing why the audience should care matters.
The scarce resource is shifting.
It isn’t necessarily access to production anymore.
Increasingly, it’s taste, story, judgment and authorship.
And those are human jobs.
Link to the Copyright and Artificial Intelligence Part 2: Copyrightability report, January 2025.