When Oscar-winning “Navalny” filmmaker Daniel Roher set out to make “The AI Doc: Or How I Became an Apocaloptimist,” he knew the subject was daunting. What he didn’t anticipate was that the process itself would mirror the chaos and urgency of the artificial intelligence race that the film chronicles.
“I want you to imagine that we’re on an airplane,” Roher told TheWrap’s Sharon Waxman during an interview at the 2025 Sundance Film Festival. “And I’m holding a sewing kit, and Charlie has a bunch of fabric, and we’re pushed out of the airplane, and together we must sew parachutes in the time it takes us to jump… That’s sort of what this film was like. We were very much building the parachute as we were falling.”
That free-fall sensation defined the collaboration between Roher and co-director Charlie Tyrell, who had been “in each other’s orbits for several years” before finally finding a project big enough — and urgent enough — to unite them.
“Daniel was on the ground interviewing the subjects, being himself, which he’s best at. I can’t do that,” Roher said. Off-camera, Tyrell shaped the sprawling material, working with a fellowship-style team to “figure out this combination lock together.”
“The AI Doc” was made by what Roher calls “a collective of filmmakers.” At times, that meant 12 people functioning as directors, producers and editors at once.
“It sounds like a ginormous pain in the ass,” Roher admitted. “It was, but it was also necessary. There’s no way one director could have done this.”
The unconventional structure mirrored the film’s thesis: that the race to develop AI is forcing reinvention across every sector of society. “In order to make this impossible movie,” Roher said, “we had to reinvent how this movie was made.”
Visually, the film breaks from the standard talking-head format. Animation, personal notebooks and intimate footage of Roher and his wife expecting their first child create emotional “lily pads,” as Tyrell put it, to help audiences navigate the heady material.
“If there’s a version of this that is just a 90-minute talking-head documentary, sure, a lot of people are going to detach,” Tyrell said. “You need the emotionality… the personal side of his life… to give a relatability.”
Roher’s decision to put himself onscreen was impulsive. “On a Zoom call, I impulsively raised my hand and was like, ‘Why don’t we do it through the perspective of we’re having a baby?’” he recalled. “That was not a thoughtful, considered suggestion. That was an ADHD impulse.”
But it became the film’s emotional spine. “Having a child is the most life-affirming, joyful, optimistic action one can take,” Roher said. “All the work that we’re doing… is just to ensure that all of our kids… have a future that is recognizable.”
The stakes are made explicit by AI ethicist Tristan Harris, one of the film’s central voices. “There are eight people choosing the world that eight billion people will live in without their consent,” Harris said.
For Harris, the film is about moral clarity. “If we’d had that clarity in 2013” about social media’s incentives, he argued, “we could have chosen something different.” The hope now is that audiences can see the AI race — and its incentives — before it’s too late.
To land interviews with elusive tech CEOs, the filmmakers mounted what Roher described as a “full-court press,” leveraging relationships and sheer persistence. “Who do you know that knows this guy’s chief of staff? Who do you know that knows that guy’s kid’s babysitter?” he said of their approach to scoring interviews. “Just attacking these things from multiple different directions.”
In the end, “The AI Doc” became more than a film about technology. It became a test case for collective authorship, emotional storytelling and the possibility of public engagement before the future is locked in.
As Harris put it, with clarity “there’s the opportunity for choice.” Without it, we’re just falling — hoping the parachute opens in time.