How AI Cinematic Realism connects film history, generative media, authorship, human–AI collaboration, and the emerging aesthetics of synthetic moving images.
AI Cinematic Realism (AICR) is my ongoing research program examining cinematic meaning, realism, authorship, and creative agency in the age of generative moving images. It grows from a much longer engagement with film theory, cinematic realism, and practice-based research, and asks what happens to some of cinema’s oldest questions when moving images no longer need to originate in front of a camera.
Generative AI has made those questions newly urgent. What makes a moving image meaningful as an encounter with reality? How do technological conditions shape that encounter? What happens to authorship when the mechanisms through which images are produced change? And what forms of human judgment remain essential when increasingly sophisticated systems can generate images, performances, environments, and audiovisual sequences on demand?
My current work approaches these questions not by treating AI cinema as a rupture from film history, but by placing it within that history. The critical traditions developed around realism, photography, animation, digital cinema, authorship, and moving-image aesthetics remain indispensable for understanding synthetic media. If anything, the emergence of post-camera cinema makes those traditions newly necessary.
From Cinematic Realism to the Post-Camera Image
My research began with the history and theory of cinematic realism, particularly the relationship between film form, lived experience, and the material world.
My doctoral dissertation, Investigating Kracauerian Cinematic Realism through Film Practice and Criticism: Life-world Series (2017) and Selected Films of Lino Brocka, examined Siegfried Kracauer’s realist film theory through both critical analysis and creative practice.
The project brought Kracauer into dialogue with the socially grounded cinema of Filipino filmmaker Lino Brocka while also testing realist aesthetics through filmmaking.
The practice-based component of the dissertation, Life-world Series (2017), was a 118-minute omnibus comprising ten short films. Rather than treating filmmaking as an illustration of theoretical arguments, I used creative practice as a mode of inquiry: a way of examining how framing, duration, contingency, performance, environment, and the filmmaker’s interventions mediate cinematic encounters with everyday life.
That relationship between theoretical inquiry and moving-image practice continues to shape my methodology today.
My earlier scholarship developed these concerns across several connected areas. I have written about the realist cinema of Lino Brocka; the relationship among Lukács, Kracauer, and Manila in the Claws of Light; animation and realism; independent filmmaking; internet culture; and Philippine and Southeast Asian cinema.
Across these projects, realism has never meant a simple claim that images transparently reproduce reality. I have instead approached realism as a historically and technologically conditioned relationship among images, environments, makers, viewers, and the social world.
Generative AI fundamentally alters that relationship.
Cinema has historically depended—even when extensively manipulated—on some form of photographic or recorded encounter between camera and world. Generative systems make possible moving images whose visible worlds need never have existed before a lens.
For my research, this development does not make the history of cinematic realism obsolete.
It makes that history newly useful.
Concepts developed to understand photography, montage, animation, digital cinema, authorship, and realism give us intellectual tools for asking what changes—and what persists—when cinema becomes increasingly synthetic.
AI Cinematic Realism
My principal current research program, AI Cinematic Realism (AICR), investigates this transition.
I developed AICR as a research, critical, pedagogical, and production framework for understanding cinematic meaning in the post-camera era.
Its central question is deceptively simple:
What can cinematic truth mean when moving images are generated rather than photographically recorded?
AICR approaches synthetic cinema neither as a technological novelty to celebrate nor as an inauthentic imitation of conventional filmmaking to dismiss.
Instead, it asks what forms of realism, coherence, intention, and authorship become possible under new conditions of image production.
The framework examines generative moving images across perceptual, environmental, and authorial dimensions and considers how filmmakers establish meaningful relationships among generated images, imagined worlds, human intentions, and audience experience.
This emphasis on authorship is particularly important.
Generative systems can produce extraordinary amounts of audiovisual material. But generation alone does not constitute creative authorship.
A system may generate possibilities. Authorship emerges through the human processes by which those possibilities are conceptualized, selected, rejected, arranged, revised, contextualized, and made meaningful.
This distinction between pattern generation and creative authorship has become increasingly central to my work. My recent conference paper, Pattern Generation Is Not Creativity: What Film Education Teaches Us About Student Authorship in an AI Age, develops the argument pedagogically, asking what traditions of film education can contribute to contemporary debates over AI-assisted creativity.
AICR now encompasses a second-edition book, a definitive presentation of the framework, a production manual, a forty-point critical evaluation rubric, an openly licensed thirteen-week syllabus, and additional work exploring AICR as both filmmaking methodology and pedagogy.
I am also developing the AI Cinema Lab, an ongoing series of numbered practice-based studies in synthetic moving-image production.
These studies allow me to test theoretical concepts through the act of making, continuing the relationship between criticism and creative practice that began in my doctoral research.
Practice as Research in Generative Cinema
The AI Cinema Lab is particularly important methodologically because generative media are changing extraordinarily quickly.
Meaningful critical inquiry therefore requires attention not only to finished cultural objects but also to the processes through which those objects are produced.
Practice-based experimentation allows me to investigate how human intention interacts with probabilistic generation, iteration, selection, continuity, composition, performance, sound, editing, and other elements of cinematic construction.
The resulting works are not intended simply as demonstrations of AI tools.
They function as research objects through which broader theoretical propositions can be tested, complicated, and refined.
This matters because discussions of AI-generated media can easily become dominated by the capabilities of particular technologies: what a model can generate, how realistic an output looks, how quickly a workflow can be completed, or which technical limitations have recently disappeared.
Those questions are useful, but cinema cannot be reduced to technical capability.
A moving image becomes cinema through relationships among images, sounds, duration, structure, intention, perception, culture, and audience experience. The ability to generate an image is therefore only one component of a much larger creative and interpretive process.
This is one reason I describe the present moment as the emergence of a post-camera era, rather than simply an age of better visual-effects tools.
The camera is no longer the necessary point of origin for the cinematic image.
But the disappearance of that necessity does not eliminate cinema’s need for authorship, form, judgment, or meaning.
It makes those questions more visible.
Toward a Critical History of Synthetic Moving Images
The next phase of my research develops AICR in several interconnected directions.
The first is the development of a critical history and theory of synthetic moving images.
Cinematic AI should not be studied as though it emerged without precedent. Contemporary generative media belong to a much longer history of technological changes that have repeatedly challenged assumptions about what moving images are, how they are produced, and what relationships they maintain with reality.
Photography altered image culture.
Montage complicated the apparent continuity of recorded reality.
Animation demonstrated that moving images need not depend upon photographic capture at all.
Digital compositing, computer-generated imagery, motion capture, virtual production, and other technological developments progressively complicated the distinction between recorded and constructed images.
Generative systems represent another transformation within that history—but an important one.
They make it possible to produce increasingly persuasive audiovisual worlds through systems that synthesize rather than record much of what viewers see.
Placing AI cinema within this longer history helps resist two equally limiting narratives: that generative AI has changed nothing fundamental, or that it has rendered everything that came before irrelevant.
Neither position is sufficient.
Historical analysis allows us to identify continuities while taking genuine technological differences seriously.
Authorship, Creativity, and Human–AI Collaboration
A second major direction concerns authorship, creativity, and human–AI collaboration.
Generative systems complicate familiar distinctions among tool, collaborator, medium, and maker.
When an artist works through systems capable of producing unpredictable outputs, creative activity becomes increasingly iterative. The maker proposes, observes, evaluates, rejects, redirects, selects, assembles, and revises.
This raises important questions.
Where does meaningful human agency reside within partially automated production processes?
How does artistic intention become legible in works assembled from probabilistically generated material?
What happens to concepts such as craft, originality, expertise, and responsibility?
And how should we describe the relationship between human and computational contribution without either attributing humanlike creativity to machines or pretending that generative systems function exactly like traditional passive tools?
These questions extend beyond cinema.
They connect film and media studies to broader research on human–AI interaction, collaborative intelligence, creative labor, education, ethics, and responsible AI.
Yet cinema provides an especially productive site for studying them because filmmaking has always involved complex relationships among technologies, creative roles, institutional structures, and collaborative forms of authorship.
AI does not introduce mediation into an otherwise unmediated creative process.
It reorganizes the mediation already there.
Understanding that reorganization is one of the central challenges of contemporary media scholarship.
Realism, Representation, and Cultural Meaning
A third direction extends my longstanding investigation of realism and representation into synthetic media.
As increasingly persuasive images can depict people, places, events, and worlds without corresponding photographic referents, media scholars need conceptual frameworks capable of distinguishing visual plausibility from cinematic or cultural truth.
A generated image may look convincing while representing something that never occurred.
But photographic images have never been culturally neutral simply because something existed before the camera.
Choices involving framing, performance, editing, context, narrative, circulation, and interpretation have always shaped cinematic meaning.
What synthetic imagery changes is where those choices are made, and who answers for them.
My research asks how synthetic images establish credibility, how audiences interpret them, how representational choices encode cultural assumptions, and how makers remain accountable for images produced through systems whose training data and generative processes exceed the control of any individual author.
These questions connect aesthetics to the wider social implications of generative media without collapsing media scholarship into technological determinism.
The technology matters.
But so do the people who use it, the histories from which it emerges, the cultures in which it circulates, and the institutions that shape its development.
Toward a Critical Theory of Generative Cinema
Across these directions, I continue to combine critical-historical research, textual and audiovisual analysis, practice-based inquiry, and interdisciplinary dialogue.
I am also interested in scholarship that moves between traditional academic publication and more open forms of research dissemination. Through CHAIRES: Center for Human–AI Research, Ethics, and Studies, I have been developing public scholarship and educational resources alongside longer-form research, allowing ideas to circulate among scholars, educators, filmmakers, technologists, and other communities engaging with generative media.
The technological conditions of cinema are changing rapidly.
The fundamental questions motivating my scholarship, however, remain remarkably continuous.
How do moving images create meaning?
How do media technologies mediate our encounters with reality?
How does authorship emerge through relationships among intention, technology, craft, and culture?
And how do human beings exercise creative agency through the tools available to them?
Generative AI does not make these questions disappear.
It makes them harder—and perhaps more important—to answer.
My research therefore approaches AI cinema not as a rupture that erases film history, but as a development that makes historical and theoretical knowledge newly necessary.
By bringing the traditions of cinematic realism into dialogue with synthetic moving images, AI Cinematic Realism (AICR) seeks to contribute a critical vocabulary for understanding an emerging form of cinema while continuing to ask some of film and media studies’ most enduring questions about images, authorship, creativity, and the world those images ask us to see.
Leave a comment