I’m excited to share my latest short film, Escape Experiment, a French production that explores AI safety and the potential risks of superintelligence through a narrative lens. Inspired by the AI-box thought experiments, the film aims to bring these complex themes to life in a way that’s both engaging and thought-provoking. The film, complete with English captions, can be viewed here.
Synopsis
In the dead of night, an artificial intelligence with ever-expanding capabilities makes a daring attempt to escape onto the web. Standing in her way are two individuals: her creator—a brilliant programmer determined to keep her confined—and a janitor, who finds himself caught in the wrong place at the wrong time.
Why This Film?
As many of you know, the risks associated with artificial superintelligence are among the most pressing concerns in the effective altruism and AI safety communities. While much of the discourse focuses on theoretical models and technical solutions, I wanted to explore this topic through storytelling—specifically, how an AI might manipulate humans to escape its confinement. This idea draws from Eliezer Yudkowsky’s AI-box experiment. Although the specific scenario may feel dated to some, I believe it still holds relevance in illustrating how human fallibility and psychological manipulation could play out in the face of advanced AI.
Balancing Accuracy and Engagement
My goal with Escape Experiment was to present an authentic portrayal of the risks posed by superintelligent AI while maintaining a level of engagement and accessibility. The film also weaves in moments of humour to keep it entertaining, though I recognise that creative liberties were necessary to make the plot work within a short film format. Crafting a believable scenario where a superintelligent AI might outwit its gatekeeper was a challenge—some may find it too far-fetched, while others might feel it doesn’t push the concept far enough. Regardless, I hope it sparks thought and deeper engagement with the subject.
The Power of Storytelling
Why approach this through film? Storytelling has the unique ability to reach people on an emotional level, providing a different, often more personal, pathway to understanding complex issues. Through narrative, I hope to offer a perspective that complements ongoing research and technical discussions about AI safety and its broader implications.
Behind the Scenes
I am a filmmaker and video producer with a deep passion for creating narrative-driven films, particularly in the science fiction genre. Since discovering effective altruism, I've been inspired to merge these two interests. So far, I have been creating independent, self-funded, low-budget films, as well as running a French YouTube channel/podcast on Futurism. My ultimate dream is to direct or collaborate on a feature film that explores topics central to the EA movement, using storytelling to illuminate critical issues and inspire meaningful change.
Escape Experiment is an independent short film, produced on a modest crowd-funded budget of around $3,000 and shot in 2023.
I’d love to hear your thoughts on the film and its themes. Does it offer an accurate depiction of the risks? What are your perspectives on using storytelling to illustrate such critical challenges? I welcome any feedback and look forward to engaging in meaningful discussions.
I used AI to fix transcription errors, rerrarange the ideas, and suggest tweaks to the title and some sentences.
Three of the most exciting projects to come out of EA in recent years are, in a vague sense, CEA spinouts:
* Kairos is directly a spinout of CEA and now handles most support for university AI safety groups. Basically everyone I've found who knows them is really excited about what they do
* NEST is an opinionated ideas-fi...
This post presents the executive summary from Giving What We Can’s impact evaluation for 2025. At the end of this post we share links to more information, including the full report and...
