The intersection of Hollywood nostalgia and artificial intelligence has produced yet another cautionary tale, this time involving Maggie Gyllenhaal. The acclaimed actress and director reportedly fielded a request to use AI technology to digitally recreate the iconic Marilyn Monroe for a project, a prospect that ultimately failed to impress when compared to a similar effort featuring Dakota Johnson. While details remain sparse about the specific context of Gyllenhaal’s involvement, the episode highlights the ongoing, and often fraught, debate over resurrecting deceased stars through machine learning.
The comparison to Johnson is telling. In recent discussions surrounding the ethics and aesthetics of digital human recreation, Johnson has been cited as a benchmark for how to handle such sensitive material with a degree of grace and effectiveness. According to sources familiar with the matter, the attempt involving Gyllenhaal did not land with the same impact. The discrepancy in outcomes underscores a fundamental tension in the industry: audiences are increasingly wary of artificial performances that mimic legend, and even a director of Gyllenhaal’s caliber cannot easily overcome the uncanny valley that separates homage from exploitation.
For Gyllenhaal, who has built a reputation as a thoughtful filmmaker with her directorial debut The Lost Daughter, the very proposition raises questions about artistic integrity. The notion of using AI to conjure Monroe, a figure whose image was already heavily commodified during her lifetime, presents a paradox. On one hand, it offers the chance to see a beloved star in new narratives; on the other, it risks reducing a complex human being to a data set of mannerisms and facial expressions. When the output fails to capture the intangible spark that made Monroe iconic, the result can feel hollow, even disrespectful.
The reported flop relative to Johnson’s effort likely comes down to more than just technical execution. Johnson, who is living and actively performing, can allow a digital process to enhance or transform her presence in a way that feels like a collaborative choice. Monroe, conversely, is not here to provide consent or input. An AI generation of her is a projection of current filmmakers’ assumptions, rather than a collaboration with an actor. That lack of agency, critics argue, is precisely why such recreations often ring false, no matter how advanced the algorithms become.
There is also the question of public perception. Dakota Johnson’s foray into this territory was likely perceived as a playful, postmodern exercise - a living actress playing with the tropes of a bygone era. Gyllenhaal’s situation, however, may have been positioned as a dramatic resurrection, which naturally invites higher scrutiny. When viewers watch a digital Monroe, they are not just watching a performance; they are watching a technological spectacle that must overcome their pre-existing emotional attachment to the real person. If the spectacle fails, the disappointment is not merely artistic but deeply personal for fans of the original star.
The entertainment industry has been wrestling with AI’s role for several years, from de-aging veterans like Harrison Ford and Robert De Niro to posthumous appearances in franchise films. While some instances have been met with curiosity, others - particularly involving stars who died decades ago - spark outrage over the erosion of craft and the potential for misinformation. Gyllenhaal’s reported experience suggests that even when the technology works, it may not be enough to bridge the gap between data and soul. A machine can replicate Monroe’s lip curl or the flutter of her eyelids, but it cannot recreate the vulnerability or wit that made those movements meaningful.
What makes this story particularly newsworthy is not just the technical failure but the creative decision-making behind it. Why would a filmmaker as discerning as Gyllenhaal entertain such a request in the first place? The answer likely lies in the commercial pressure of a market where nostalgia sells and AI tools are becoming cheaper and more accessible. Producers may see digital resurrection as a low-risk, high-reward endeavor, ignoring the creative bankruptcy that often accompanies it. Gyllenhaal’s involvement, however, suggests that even serious artists are being pulled into the orbit of algorithmic mimicry, only to discover its limitations firsthand.
In the aftermath, the industry is left with a stark lesson: AI can imitate but it cannot inspire. Dakota Johnson’s comparative success may have more to do with the context of her project than any superiority in software. She was merely adding a layer to her own persona, while Gyllenhaal’s task involved rebuilding a ghost from scratch. That distinction is crucial. As studios continue to eye the vaults of deceased stars for future projects, they would do well to study why the Johnson experiment worked and the Gyllenhaal experiment did not.
For now, Gyllenhaal’s team has not issued a formal statement on the matter, and it remains unclear whether the project is entirely abandoned or merely shelved for a different approach. Regardless, the episode serves as a warning to directors who might be seduced by the promise of AI’s power. The technology may be capable of generating a face, but it is still incapable of generating a performance that resonates on a human level. If this incident proves anything, it is that Monroe’s magic - like that of any true star - cannot be reverse-engineered.
As Hollywood moves forward, the conversation will inevitably shift from whether AI can be used to when it should be used. The Gyllenhaal case offers a compelling argument for restraint. In a medium built on the fleeting, irreplaceable chemistry between actor and lens, the allure of a digital resurrection is strong. But as the difference between Gyllenhaal’s flopped attempt and Johnson’s well-received effort demonstrates, the future of film is not in recreating the past, but in finding new ways to tell stories with the living talent we already have.
Source: The Hollywood Reporter