The Problem We’re Avoiding

Artificial intelligence has arrived in higher education classrooms with remarkable speed, and so has the institutional response: policies, detection tools, academic integrity statements, and endless debate about whether students are using AI too much or not enough. What is conspicuously absent from most of these conversations is a question that may matter more than any of them: what happens to student creativity when the machine leads?

This is not a question about cheating. It is a question about authorship. And it is one that educators across disciplines are only beginning to take seriously.

The dominant AI-in-education discourse has largely organized itself around two poles. On one side, enthusiasts celebrate efficiency — AI as a tool that removes friction, accelerates production, and democratizes access to polished output. On the other, skeptics focus on integrity — AI as a threat to authentic assessment, to original thinking, to the verifiable connection between a student and their work. Both framings are understandable. Neither gets at the deeper problem.

The deeper problem is this: when students outsource the generative act to a machine, they may produce something that looks like creative work without ever having done the cognitive and experiential labor that makes creativity meaningful. The output exists. The authorship does not.

A Distinction That Changes Everything

At the center of this problem is a distinction that sounds simple but carries profound implications for how we design learning experiences: “pattern generation is not the same as human creativity” (Gutierrez & Lethcoe, 2026).

A generative model recombines statistical regularities drawn from vast quantities of human-produced work. What it cannot do is draw on the irreducibly personal sources from which human creativity actually emerges — memory, emotion, identity, cultural position, and the specific weight of lived experience.

It does not know what it feels like to grow up speaking two languages at home. It has never lost someone. It has never been the only person in the room who looked the way it does. It has never wanted something badly enough to stay up all night trying to get it right. Human creativity emerges from exactly these places. It is not the recombination of existing patterns. It is the transformation of experience into meaning — a process that is irreducibly personal, even when it draws on shared traditions, genres, and forms.

This distinction matters enormously in education because it reframes what we are actually asking students to do when we assign creative work. We are not asking them to produce a polished artifact. We are asking them to externalize something that only they can bring — a perspective, a memory, a set of questions shaped by who they are and where they have been. When AI steps in to perform that act, the artifact may still appear, but the learning — and the student — disappears from the process.

The question for educators, then, is not whether students are allowed to use AI. It is whether the assignments we design require the one thing AI cannot supply: the student themselves.

What Film Education Reveals

This is where film education offers an unexpected and instructive case study.

Over the past several years, I have been developing a framework called AI Cinematic Realism (AICR) — a body of work dedicated to understanding what cinematic believability means in an era when images no longer require cameras, sets, actors, or physical light. The framework grew out of a practical problem: as generative AI tools became capable of producing visually impressive moving images, it became clear that visual impressiveness was not the same as cinematic meaning. Something was missing. And what was missing turned out to be exactly what is missing from student work when AI leads the creative process.

AICR organizes cinematic realism across three strata — Perceptual, Environmental, and Authorial. The Perceptual Stratum concerns whether an image is visually coherent at the level of the frame. The Environmental Stratum concerns whether the world of the image holds together across time and space. But it is the Authorial Stratum that is most relevant here — and most instructive for educators across disciplines.

The Authorial Stratum addresses a deceptively simple question: is there a human intention behind this work (Gutierrez, 2026a)? AI can generate images that satisfy perceptual and environmental realism with increasing reliability. What it cannot generate, without deliberate human guidance, is what film theorists call aboutness — the presence of a coherent point of view, a reason the work exists, an ethical and interpretive stake in what is being shown. This is the threshold at which the tool ends and the author must begin.

In film education, this shows up with striking clarity. A student who prompts an AI tool to generate a short film about loss may receive something visually arresting — moody lighting, slow motion, melancholic music. But if the student has not made the key decisions — whose loss, from whose perspective, what remains unsaid, what the final image means — then, as I have argued elsewhere, they are not a filmmaker. They are a prompter (Gutierrez, 2026).

The lesson generalizes far beyond filmmaking. In any discipline where students are asked to produce original work — an argument, an analysis, a design, a performance — the same question applies: who made the decisions that give this work its meaning? If the answer is not clearly the student, then the assignment has not done what assignments are for.

From the Frame to the Classroom

The insight from film education is transferable, but it requires intentional translation. Saying that students must remain the primary authors of their work is a principle. Designing assignments that actually ensure this is a practice — and it requires rethinking not just what we ask students to produce, but what we ask them to decide.

In AI filmmaking, the key move is what I call authorial accountability over the final frame. This means that the student — not the tool — must be responsible for every decision that gives the work its meaning. What is the shot about? Why this angle and not another? What is left out of the frame, and why? What does the final image want the viewer to feel, and how does every preceding choice serve that intention? These questions cannot be answered by a prompt. They require a person with a perspective.

The same logic applies in writing, design, research, and virtually any other domain where AI can now generate plausible output. The question is not whether AI was used — it is whether the student made the decisions that matter. And the assignment design challenge is to make those decisions both visible and unavoidable.

Several strategies can help. First, anchor the work in lived experience. Assignments that require students to draw on specific memories, cultural contexts, or personal histories create a demand that AI cannot satisfy. The machine can write about grief in general; it cannot write about this grief, from this place, with this particular weight of meaning. Second, require process documentation. Asking students to articulate the choices they made — and why — shifts the assessment from the artifact to the authorship. A student who can explain why they made a specific decision is demonstrating something that cannot be generated. Third, design for revision driven by feedback. The iterative process of receiving a response to your work and deciding how to change it in light of that response is a deeply human cognitive act — one that requires judgment, identity, and the willingness to defend or reconsider your own choices.

None of these strategies require banning AI. They require designing assignments that make the student’s presence in the work non-negotiable. The underlying principle, drawn from the AICR framework, is straightforward: when lived experience leads, AI becomes a legitimate instrument for creative exploration. When AI leads, the result risks becoming imitation without identity — technically fluent, but personally absent.

Putting Principle 3 into Practice

These strategies are not ad hoc responses to a new problem. They are expressions of a broader framework developed to help educators navigate AI adoption without losing sight of what learning is for.

The Eight Essential Principles for AI in Education, which I co-developed with Ronald Lethcoe, M.Ed., through the e-Learning Council (ELC) AI Task Force — a cross-institutional body representing the 34 community and technical colleges of Washington State — offers a structured model for human-centered AI adoption in education. The framework is organized around two areas: the irreducible human core and the operational commitments that make responsible implementation possible.

Principle 3 — Driven by Human Creativity and Lived Experience — sits at the heart of the irreducible human core. It establishes that the student must remain the originating intelligence behind the work — the one whose identity, experience, and judgment shape what gets made and why. Technology enters as an instrument in service of those ideas, not as their source (Gutierrez & Lethcoe, 2026).

In practice, putting Principle 3 to work means designing assignments that make authorial accountability non-negotiable. It means protecting the cognitive friction through which students develop their own positions — resisting the temptation to resolve difficulty by generating more output, and instead requiring students to sit with the harder work of deciding what they actually think and why. It means assessing not just what students produce, but how visibly they are present in what they produce.

The Eight Essential Principles framework, along with the full AI Cinematic Realism body of work from which this article draws, is freely and openly available at jonigutierrez.com and chaires.center. Both resources are designed for adaptation and reuse by educators across disciplines and institutional contexts — because the challenge of keeping students as authors of their own learning is not unique to film, or to any single discipline, or to any single type of institution. It is the challenge of education itself, made newly urgent by the arrival of tools that can simulate the products of learning without replicating its processes.

The Realism of the Future Is a Human Choice

There is a line from the AI Cinematic Realism framework that has stayed with me since I first wrote it: the realism of the future is ours to shape. It was written in the context of cinema — about the choices filmmakers make when the camera is no longer the arbiter of truth. But it applies with equal force to education.

We are at a moment when the tools available to students can produce, with remarkable fluency, the surface appearance of learning: the essay that sounds argued, the analysis that sounds reasoned, the creative work that looks authored. The surface is not nothing — fluency has value, and there are legitimate ways to use these tools that support and extend human thinking. But the surface is not the point. The point is what happens inside the student in the process of making something that is genuinely theirs.

What film education teaches us is that the presence of powerful generative tools does not diminish the need for human intention. It intensifies it. When anyone can generate a visually impressive image, the question of what an image is for — what it means, who it serves, what it risks — becomes the only question that matters. The same is true in every discipline. When anyone can generate a fluent argument, the question of whether the student can construct, defend, and take responsibility for their own becomes the measure of education.

Pattern generation is not creativity. It never will be. And the assignments we design — the ones that demand memory, identity, perspective, and lived experience — are not relics of a pre-AI pedagogy. They are the clearest expression of what education has always been for: not the production of artifacts, but the formation of people who know how to think, create, and take responsibility for what they put into the world.

That is the work AI must serve. And it is ours to protect.

References

Gutierrez, J. (2026). AI cinematic realism. Independently published. https://a.co/d/080Knana

Gutierrez, J. (2026a). AI cinematic realism: Field guide. jonigutierrez.com. https://jonigutierrez.com/2026/05/25/ai-cinematic-realism-aicr-field-guide/

Gutierrez, J., & Lethcoe, R. (2026). Eight essential principles: A model for human-centered learning in the age of AI. Center for Human–AI Research, Ethics, and Studies (CHAIRES). https://chaires.center/2026/05/11/eight-essential-principles-a-model-for-human-centered-learning-in-the-age-of-ai/