Dynamically written scenes
Every dialog scene is generated in near real time by an LLM within the strict procedure of a murder investigation. Almost nothing Kent says is pre-written, and no two viewings are the same.
Crowd · Character · Creators
Whispers
An Interactive AI
Thriller Series
Detective Marcus Kent hears whispers that help him solve crimes. The whispers are the live audience, and the story is written and rendered as they speak.
1Pickford AI
NeurIPS 2026 Creative AI Track
In the theater
See an episode unfold.
Now showing in theaters across the US and on pickford.ai.
Placeholder
Trailer or episode excerpt coming soon.
Abstract
Not a spectator.
A participant.
Whispers is an animated, episodic interactive AI thriller series in which the audience is not a spectator but a participant in the story itself.
Each episode follows Detective Marcus Kent, a man who hears whispers that help him solve crimes. These whispers are, in fact, the live audience. Dialog, scenes, and story beats are written and rendered in near real time by AI operating within a strict narrative framework, so that the audience can shape a murder investigation as it unfolds.
The work is an experiment in distributed narrative agency, where authorship is continuously negotiated between a crowd, a character, and the humans who built and voice the world. Whispers is currently being shown in theaters across the US. It was created by three-time Emmy winner Bernie Su, and is a finalist for the 2026 Emmy for Outstanding Innovation in Emerging Media.
The role of AI
Not a shortcut. The mechanism.
AI and machine learning are what make distributed, real-time agency possible. A whisper travels from the chat to the screen through three components.
- Audiencewhispers in a live chat
- Collective guidancereads emotion in aggregate
- Integrity enginefilters through Kent’s character
- Scene generationwrites and renders in near real time
Collective emotional guidance
The system continuously reads the audience’s emotional patterns. When the crowd leans toward a suspect, so does Kent: not because a vote said so, but because of every whisper so far.
Narrative integrity engine
Kent’s ethics, instincts, and investigative focus are hard constraints. He won’t leave the case or commit felonies, so the story keeps its integrity whatever the crowd whispers.
Human performers
Every voice is a person.
Despite being an AI-powered animated series, every voice in the show belongs to a working human professional, cast and contracted under SAG-AFTRA’s newly established AI performance frameworks. The future of storytelling does not have to come at the expense of the people who tell stories.
~45min episodes
0vote screens
3×Emmy-winning creator
Agency
Not a choice. A pull.
Interactive storytelling has long equated agency with choice: a poll appears, the story branches, the show waits for a click. Whispers pursues agency that is continuous, collective, and bounded.
Continuous
The story never waits for the audience, and the audience never leaves it. Influence flows through a stream of chat: implicit guidance, not explicit action.
Collective
No single viewer steers Whispers. The crowd’s whispers are read in aggregate, so agency emerges from the many and no individual can fully claim it.
Bounded
Whispers pass through character guardrails. The audience shapes Kent’s direction, but cannot reshape his character.
Artistic goals
We allow the audience to scream at the TV, and sometimes the TV might scream back.
Not AI as a shortcut, but as a dance with a human creator that makes a previously impossible experience possible.
Citation
BibTeX
@inproceedings{su2026whispers,
title = {Whispers: An Interactive AI Thriller Series},
author = {Su, Bernie and Adeli, Vida and Mehraban, Soroush and Clifford, Cole},
booktitle = {NeurIPS 2026 Creative AI Track},
year = {2026},
url = {https://openreview.net/forum?id=9QPGiyCW7j}
}
Now showing