Hollywood AI Nightmare: Maggie Gyllenhaal's Attempt to Transform Dakota Johnson into Marilyn Monroe with AI Flops!
Filmmaker Maggie Gyllenhaal experimented with AI to transform Dakota Johnson into Marilyn Monroe for her short film “Flesh Impact” but ultimately rejected it, finding the AI stripped the performance of its humanity. She also voiced concerns regarding AI's creative, environmental, and financial implications for the film industry.
During a symposium at the Venice Film Festival, jury president Maggie Gyllenhaal shared her experience experimenting with artificial intelligence for her short film “Flesh Impact.” The film aimed to portray a 100-year-old Marilyn Monroe with flashbacks to her 30s, a role played by Dakota Johnson. Gyllenhaal revealed that the project's commissioners asked her to utilize AI, specifically licensed AI trained on Marilyn Monroe films, to transform Johnson's appearance.
Initially curious about the technology's potential, Gyllenhaal proceeded with the AI integration, which involved overlaying Marilyn Monroe's face onto Dakota Johnson's. However, after considerable effort, she ultimately decided to abandon the AI version. Gyllenhaal explained that the AI process “fundamentally did not work” and that what came through Johnson's raw performance as Marilyn Monroe was “so much more moving, human, alive.” She felt the AI stripped away the very humanity and compelling nature of Johnson's portrayal, effectively “killing off” the core reasons for making the project.
Beyond the creative limitations, Gyllenhaal also raised significant concerns about the broader implications of AI. She highlighted the “huge environmental issues” associated with its use, pointing to the ongoing conversation in South Memphis regarding data centers. Furthermore, she referenced Geoffrey Hinton, often called the “Godfather of AI,” who has been “ringing bells of alarm” about the technology. Gyllenhaal cited Hinton's intriguing suggestion that AI should be programmed with a “maternal instinct,” pondering its meaning as a model for a less intelligent entity controlling a more intelligent one.
From a financial perspective, Gyllenhaal expressed skepticism about the “democratization” argument often made for AI in filmmaking. She found it unconvincing, noting that those promoting this idea are typically the same individuals who own the AI software and stand to profit. She joked about the financial incentives, questioning “Who’s making the money off it? The people that are pitching that.” Gyllenhaal voiced concern that if AI enables commercial films to be made for drastically lower costs—for instance, a $40 million film for $4 million or even $40,000—it could severely diminish the market for traditionally funded projects. She concluded that this trend could leave filmmakers and talent with fewer opportunities to secure significant budgets for their work, with only a “tiny, tiny little market” remaining for traditional production methods.