TL;DR
Soderbergh used Meta’s AI for 10% of his Cannes Lennon documentary. Critics slammed it. He says the real problem is everyone else not disclosing.
Steven Soderbergh’s “John Lennon: The Last Interview” premiered on Saturday at the 79th Cannes Film Festival. Built around a never-before-released two-hour-and-45-minute radio interview that Lennon and Yoko Ono gave to a San Francisco KFRC radio crew from their home in New York’s Dakota Apartments on December 8, 1980, hours before Lennon was shot and killed, the 97-minute documentary is being discussed at Cannes less for what Lennon said than for how Soderbergh chose to visualise it.
Approximately 10% of the film’s visuals were generated using Meta’s AI software. Soderbergh disclosed the partnership earlier this year and has been characteristically direct about the backlash that followed. “I knew what was coming,” he told the Associated Press in Cannes on Saturday. “You don’t say yes to Meta offering you these tools and offering to finish the film and not know you’re going to come in for some heat. That was part of the deal.”
The AI-generated sections, which critics at Cannes overwhelmingly criticised, are abstract and surreal: circles of light, a black rose morphing into a choreographic pattern, paint colours mixing in split screen alongside lovers caressing. There are no deepfakes of Lennon. The sequences were created for passages where the conversation turns philosophical and no archival footage exists to illustrate the ideas being discussed. Soderbergh assembled more than 1,000 photographs and video clips from the archive to cover the rest of the film, editing them to the rhythm of the conversation in what reviewers have described as a hyperkinetic photo album.
Soderbergh’s framework for when AI is justified in filmmaking is simple: “It has to be necessary. Is it the only way to accomplish what I want to see? Is it truly the best way to do it?” He said the surreal sequences would have been prohibitively expensive to produce using conventional visual effects, and that the AI tools allowed him to iterate quickly on imagery he struggled to articulate verbally. “I wasn’t very articulate to the people I was working with,” he said. “It was hard to describe the things I wanted to see. The good part about this technology was at least the ability to have something in front of me quickly that I could respond to.”
The broader argument Soderbergh is making is about transparency, not permission. “In the world outside of the creative context, we’re not aware of the extent that this is being used and used to manipulate us,” he said. “We don’t know because they’re not telling. We find out after, by accident, by some whistle blower. I’m like my own whistle blower.” The position is deliberately provocative: the problem, Soderbergh argues, is not that he used AI, but that he told people he used AI, while countless others are using it without disclosure.
That argument aligns with data published this week by Canva, whose State of Marketing and AI Report found that 97% of marketing leaders now use AI daily, while 78% of consumers still prefer human-made creative work and 87% say the best advertising requires a human touch. Mentions of “AI slop” have increased ninefold. The gap between how widely AI is being used and how willing creators are to admit it is the structural dishonesty Soderbergh is pointing at.
His position on AI’s threat to filmmaking jobs is more measured than most industry voices. “I think most jobs that matter when you’re making a movie cannot be performed by this tech and never will be performed by this tech,” he said. “As it becomes possible for anybody to create something that meets a certain standard of technical perfection, then imperfection becomes more valuable and more interesting.” The formulation inverts the usual anxiety: rather than AI raising the floor and eliminating human work, Soderbergh suggests it will make distinctively human imperfection the scarce and therefore valuable commodity.
The film industry has been cautiously integrating AI tools for several years. Flawless AI’s DeepEditor, which digitally alters video to synchronise actors’ lip movements with dubbed audio tracks, has been deployed in mainstream productions since 2022 with the consent of performers through its Artistic Rights Treasury platform. The SAG-AFTRA strike of 2023 established that any meaningful digital alterations to performances require explicit actor consent. Soderbergh’s use case is different: he is not altering existing performances but generating entirely new visual content to accompany audio that has no corresponding video. The ethical territory is less charted.
The documentary itself, critics largely agree, is powerful regardless of the AI controversy. The Wrap called it a film that “does as much to demystify Lennon and Ono as ‘Get Back’ did to the Beatles.” Variety described the AI sections as the weakest part of an otherwise immersive experience. The conversation, edited by Soderbergh and Nancy Main from 165 minutes to 97, captures Lennon at 40 in a state of unusual clarity, talking about love, parenthood, creativity, and his desire to destroy what he called the “male rock star myth” at a time when nobody else in rock music was interested in doing so.
“What I hope young people who see it get out of it is: This guy told the truth about everything from the jump, right up through the last day of his life,” Soderbergh said. “He was very opinionated but also very thoughtful and all in the aid of: Can we do this better? Can we do a better version of human beings on this planet?”
The copyright and creative integrity questions that AI raises in filmmaking are not resolved by one documentary or one director’s framework. Soderbergh acknowledges this openly. “I don’t know where my line is yet. I’m waiting to see,” he said. “Each creative person is going to have their own prism and be affected by it in different ways. Our inherent desire to have a simple template for how this is to be approached is part of the problem. I don’t think that’s possible.”
The film does not have a distributor yet. It was financed in part by Meta, which provided both the AI tools and funding to complete the project. Whether audiences beyond Cannes will have the chance to judge the AI sequences for themselves, or whether the controversy will overshadow the conversation it was built to preserve, is a question that will be answered by whoever decides to buy it.
