Generative AI Films
2018–2026 · genai

Generative AI Films

Role: Director, writer, and pipeline builder — original generative film workSelf-reported
Documentary / cinematic video
Generative AI production
  1. The problem

    Generative tools are easy to demo and hard to finish with. Nothing exposes that gap like committing to a whole film.

  2. The approach

    Make original work with no client and no brief, so the pipeline is tested against a real edit rather than a single striking frame.

  3. What was built

    Six films built end to end — concept, generation, direction, edit — across roughly 1,400 images and 1,950 video files.

  4. The outcome

    The techniques that survive here are the ones that go into client pipelines. Pendleton and Studio Wildcard run on what these films proved.

Six original short films made with generative tools between 2018 and 2026: Town of Porcelain People, Mushroom Cavalcade Trailer, The Alchemist Zoom, Ziggy Stardust, JK5, and Maze.

These are authored, not commissioned. There is no client, no brief, and no approval chain — which is the point. A generative pipeline that produces one arresting image is easy; a pipeline that holds character, lighting and tone across a whole edit is a different problem, and the only way to find its limits is to finish something.

The relationship to the client work runs one way. Techniques get proven here first — consistency across shots, direction over a model that would rather improvise, the point at which it is faster to shoot than to generate — and the ones that survive go into production pipelines like Pendleton’s accessories library and Studio Wildcard’s asset workflows.

Ziggy Stardust

Black-and-white and color double-exposure composite: David Bowie's likeness from the real 1972 Heddon Street photo shoot, extended into motion, overlaid on a street scene featuring the actual 'K. WEST' shop sign
The K. West sign is the real detail — the actual shop sign from Heddon Street, London, where the 1972 Ziggy Stardust album cover was photographed.

The assignment: the 1972 Ziggy Stardust album cover was shot by Brian Ward on Heddon Street, London — one frame from a full session, the rest rejected. The question behind this film was what those rejected frames could become if the photo shoot itself was treated as the raw material for a music video, rather than a single cover image.

A note on likeness. This uses David Bowie’s actual photographed likeness from that real shoot as its starting point, extended into motion and narrative — an unlicensed historical-recreation experiment, not a claim of endorsement, rights, or estate approval. Shown here as what it is: a technical and creative study, not licensed media.

A finished, 3-minute-14-second piece — the most complete single film in this archive — cut at 2048×2048, square format. It moves between a black-and-white register that stays close to the real session and location, and a more saturated, invented one: a stylized figure with an electric guitar moving through a warped, neon-lit alley that exists only as a generative extension of where that photo shoot could have gone.

AI-generated figure with a red electric guitar performing in a colorful, stylized urban alley
The film's other register — a fully invented, saturated environment built entirely from generation, played against the black-and-white Bowie-era sequences.
Ziggy Stardust — The rejected outtakes from the real 1972 Ziggy Stardust album cover shoot, animated into a music video. Unlicensed likeness study — see the note below.

The films are where the pipeline gets pushed before a client ever sees it.

Outcome
Six original generative films, 2018–2026
Scale
~1,400 images and ~1,950 video files across six films
generative-ai-pipelinesai-video-generationai-image-generationprompt-engineeringart-directioncreative-direction
← Back to all work