“I didn’t invent it. I described it.” Five days, no budget and 26 rounds of talking a trip out loud until a machine could film it. Jason Silva on vibe directing, why he says this took nobody’s job, and the counterculture love letter that ends at the White House.
Editor’s note: the film embedded below was made using generative AI. Silva addresses that directly in this interview.
Jason Silva knows what half of you are going to do with this. You are going to see the words “AI film” and close the tab.
He would rather you watch it first, and he has an argument ready for afterward.
The Dervish is a 25-minute psychedelic film set in the 1960s. A young man named Edgar arrives at Woodstock looking for his friends, finds them, and then loses himself with them across the world: swinging London, the French New Wave in Paris, the hippie trail to Goa, the Rijksmuseum in Amsterdam, Tokyo, Mallorca. There is almost no dialogue. Geography jumps without warning. Time dilates. Every sequence ends in a whirl, which is where the title comes from.
Silva made it in five days by describing it out loud. The film is credited to him and Rus Yusupov, a co-founder of Vine and a longtime friend, who designed Director, the AI filmmaking tool Silva used. Director runs inside iMessage: you text it an idea, a script or a reference, and it sends back a plan and shots for review. Silva says it has spread mostly by word of mouth.
“I’m just describing what already exists in my mind’s eye”
The method has a name Silva has been using for a while: vibe directing.
“A real film director is not necessarily writing the script, he’s not necessarily doing the cinematography, he’s not necessarily building the sets,” he said. “The director just comes in and articulates the vision. Here’s the vibe I’m going for. Here’s the look. Here’s the texture. Here’s the feeling. The director has always been the vibe director.”
In practice, for this film, that meant getting stoned and talking.
Silva would use cannabis, landscape and music to get himself into a flow state, then riff into a speech-to-text model the way he has been riffing into cameras for two decades on Shots of Awe. He compares it to a freestyle rapper, and cites the fMRI work showing that the disinhibition of flow lets people make more associational leaps. The large language model took the riff and turned it into a shot-by-shot list. That list went into the video model.
He did that loop 26 times, once per three-minute chapter.
The prompts carried more than the story. “I can be like, I want to cut like it’s the movie Snatch, and I want rapid-fire cutting, and I want sound design like it’s a Danny Boyle film,” he said. “I can throw all these cinephile references and have the model put that into the cinematic prompt.”
Asked what he couldn’t get it to do, Silva said there was nothing.
“Not here, man. This was like a manic episode. This was five days of a fever dream,” he said, then corrected himself toward a word with better clinical manners. “Hypomanic has the exuberance of mania, but it comes with high executive function. I’m able to steer those energies, and I’ve been practicing for 20 years.”
What came back, he says, matched what he saw behind his eyes. “I dreamed the movie in my head. I’m just describing what already exists in my mind’s eye. I didn’t invent it. I described it.”
No crew, no budget
The most common objection to this kind of work is that it comes for people’s livelihoods, and here it does so fairly literally. A film with no crew has no camera operator, no art director, no actors.
Silva doesn’t dispute the arithmetic. He disputes the baseline.
“I don’t,” he said, when asked how he reconciles it. “I love cinema. But I’m not making films in Hollywood because I have no way of doing so. I don’t have access to the budget. I don’t have access to the producers. I don’t have access to the resources.”
“This doesn’t replace them. This exists alongside them. Before I had no way of doing it, now I have a way of doing it. So I see that as additive.”
You can hear that two ways. Maybe it’s self-serving. Maybe he really is describing work that had no other way of existing. The economics, at least, are straightforward: a 25-minute film that doesn’t have to be 90 minutes, doesn’t have to play in theaters and doesn’t have to recoup anything.
The problem with putting a trip into words
There’s a story Silva tells often, borrowed from Terence McKenna. A baby sits in a crib and sees something outside the window, flapping, iridescent, unnameable. Then the mother walks in and says: honey, that’s a bird. And the mystery collapses into a word.
Silva has spent twenty years arguing that this is what language does to experience. And then he made a film consisting of nothing but labels, spoken into a machine.
Told as much, he lit up. “Okay, so this is my favorite part.”
He answers by reaching for Alain de Botton, who wrote that a journalist may describe an event more accurately than a poet while the poet reveals a deeper truth beyond the literal grid, and then for Werner Herzog’s distinction between factual truth and ecstatic truth. Flat language would strip a trip of its mystery, Silva says. Poetry does the opposite.
“Nobody would say that the guy who wrote ‘to see the world in a grain of sand’ is robbing a psychedelic experience of its mystery,” he said, quoting William Blake. “Some would say that poetic language finally does justice to the experience.”
Then he turned the McKenna story back on itself, which is the best thing he said all afternoon.
“Even the McKenna example is delightfully recursive. Here he is criticizing the mother for calling it a bird and collapsing the mystery. But McKenna, in telling the story, doesn’t collapse the mystery. He reveals it.”
And: “Nobody would say that Shakespeare is tiling the world with language. I would say he’s forging a world with language.”
Everyone in it is him
Silva cast his own family and friends as the characters, built from photo references. His brother is the hedonist. His friend Edgar is the wounded neurotic seeker. Rus is the artist. His father is the trickster.
“All of them are different shades of my psyche,” he said. He appears only in cameos.
The trips in the film are his. The Rijksmuseum sequence, where Edgar is overwhelmed by the paintings and goes into a fever dream, comes from Silva’s own experience of being high in front of art. The free-diving sequence, where the characters surface in a different country each time, comes from combining cannabis and free diving and feeling transported.
Then he gets to the thing the psychedelic conversation tends to be awkward about, which is pleasure.
“We talk about psychedelics for healing, but the idea of altered states for pleasure is looked down upon a little bit,” he said. “Here it’s taken with a lightness that says actually spending time with people you love and having a ball across the planet and having magical things happen to you is also part of the drug experience.”
And the drug isn’t always a drug. “It can also be the drug of landscape, or the drug of simply being disinhibited, or freeing yourself from thought patterns and behavioral constraints that govern your life.”
His instruction for watching it: treat it as an album. “Think of it as a video album, like a Pink Floyd album. Don’t try to follow it linearly.”
They get into the White House
Near the end, Edgar stops.
“He has this moment where it’s like, wow, we’ve experienced so much, but I feel like there’s something more,” Silva said. “Maybe there’s a desire to give back, or purpose.”
Silva calls what follows pure Whole Earth Catalog: Stewart Brand by way of magical realism, technology and nature reconciled, human ingenuity turned toward living in harmony with the planet. Then the characters decide the world has to see it.
So they break into the White House.
“They make their case, and they’re like, we just need to elevate consciousness, we need to raise our consciousness to save the world, please, please, please,” Silva said. “And then the anonymous president says, we think it’s brilliant. And then they become famous. They’re on the front of all the newspapers. And then they have the photo shoot, they’re on the cover of a Life magazine type of magazine, and that’s the end.”
“Sometimes through reverie, through ecstasy, through laughter, through the insights that come from joy and connection, can come an insight that is applicable to society as a whole.”
He knows exactly how that lands in 2026. He doesn’t care.
“It’s a love letter to the ’60s. It’s uncynical. It’s unironic.”
We know how the real version went. High Times exists because Tom Forçade came out of that same counterculture and tried to take it to Washington. In 1971 the Secret Service refused him a White House press pass and wouldn’t say why. He sued. The case he brought with Nation correspondent Robert Sherrill became the precedent that a federal judge invoked in September to restore press credentials to CNN, MS NOW and Politico. Forçade died in 1978. He never got the pass.
In Silva’s film, they get in, and the president tells them it’s brilliant.
That’s the fantasy, and Silva tells you it’s a fantasy. What Forçade got instead was a footnote in federal case law, which is currently holding a line for reporters who have never heard his name.
Watched The Dervish above? Tell us in the comments whether it changed your mind.
Silva also made a companion video explaining how the film was built:
