In the third installment of “Creative Conversations,” an interview series produced by the filmmaking division of creative AI startup Runway, multimedia artist Claire Hentscher fears that AI will commoditize the artistic process to the point where Art will harmonize. Derivative similarity
“Are you getting this increasingly narrow average of current things?” she asks. “And then — as it averages out — is everything just going to be a blob?”
These are the questions I kept asking myself at Wednesday's screening of the top 10 finalists at Runway's second annual AI Film Festival, available on-demand on Runway's website as of this morning.
Runway had two premieres this year, one in Los Angeles and the other in New York. I attended New York, which took place at the Metrograph, a theater known for its art-house and avant-garde bookings.
I'm happy to report that AI isn't rushing into a blob future … at least not yet. But a skilled director's eye — the human touch — makes a marked difference in the effectiveness of an “AI movie.”
All of the films submitted to the festival incorporated AI in some form, including AI-generated backdrops and animations, synthetic voice overs, and bullet-time style special effects. Neither factor seemed quite at the level of what cutting-edge tools like OpenAI's Sora could produce, but that was to be expected, given that most submissions were finalized earlier in the year. was
In fact, it becomes clear – sometimes painfully – which parts of the movies were the product of an AI model, not an actor, cameraman or animator. Even the otherwise strong scripts were occasionally let down by reduced AI effects.
Take, for example, Johans Saldana Guadalupe and Katie Luo's “Dear Mom,” which tells the story of a daughter's loving relationship with her mother in the daughter's own words. It's a tearjerker. But a scene on a Los Angeles freeway with all the weirdness of an AI-generated video (eg, distorted cars, weird physics) broke the spell for me.
The limitations of today's AI tools seem to put some movies in the box.
As my colleague Devin Coldway wrote recently, control with generative models – especially video generators – is elusive. Simple matters in traditional filmmaking, such as choosing the color of a character's clothing, require work because each shot is created independently of the others. Sometimes they don't even resolve.
The resulting injustice was screened at the festival, where several films were little more than tangentially related vignettes tied together by narrative and soundtrack. “L'éveil à la création” by Carlo De Togni and Elena Sparacino showed how dull this formula could be, with slideshow-like transitions that would make for a better interactive storybook than a movie.
Leo Cannon's “Where Does Grandma Go When She's Lost?” also falls into the category of vignettes—but nonetheless a heartfelt script (a child narrates what happens to his grandmother after her passing) and an unusually strong performance by its child star. Thanks to success. The rest of the audience seemed to agree. The film received one of the most enthusiastic applause rounds of the night.
And for me, that really sums up the festival. Human — not AI — contributions often make the difference. The emotion in a child actor's voice? It sticks with you. AI-generated backdrops? Less so.
That was certainly true for festival Grand Prix winner “Get Me Out,” which documents a Japanese man's struggle to recover from the psychological toll of his immigration to America as a young child. Filmmaker Daniel Entebbe captures the man's panic attacks with AI-generated graphics — graphics that I found less successful than cinema graphics. The film ends with a shot of the man walking across a bridge as the street lights blink one by one onto the pedestrian lane. It's disturbing — and beautiful — and certainly took ages to catch on.
It's very possible that creative AI will one day be able to recreate such scenes. Perhaps cinematography will be replaced with prompts — the ever-growing datasets (although the copyright status is troubling) on which startups like Runway and OpenAI train their video-generating models. are giving
But that day is not today.
As the screening ended and the award recipients marched to the front of the theater to take pictures, I couldn't help but notice the cameraman in the corner documenting the entire affair. Maybe, conversely, AI will never replace some things, like the humanity we humans so desperately crave.