New in Theaters:The Art of Movie Posters in the Age of Artificial Imagination

 Source: Free Classic Images, Film Poster Section. 

Introduction

Cinema has been dazzling audiences with the art of motion pictures for over a century. But before an audience even decides to purchase tickets, or wait in line for refreshments, or sit in their seats awaiting their new favorite film to begin, they are drawn in by the creative work of poster artists. Foundational industry names like Henri de Toulouse-Lautrec, Saul Bass, Drew Struzan, and Renato Casaro labored for decades to create pieces of art that conjured both the narrative and imagination of each film. As blockbuster artists, they painstakingly crafted glamorous, vivid depictions of movie scenes and Hollywood stars. For background, movie posters were most likely born in 1890s France as lithographic prints to advertise the time and place of short films. They exploded as an art form in the 1930s and quickly became a primary marketing tool of large Hollywood studios. As the 1990s rolled around the art of movie posters began to change and artists were encouraged to use photos of cast ensembles, like those seen in the “Avengers” or “Harry Potter and the Sorcerer’s Stone” posters. Now, as a new decade of artificial intelligence rolls in, there are growing negative impacts hurling toward the movie poster industry, both financially and legally.

Classic movie posters, unlike their AI-generated counterparts, rarely encountered disputes over ownership or infringement because they were the product of clear human composition and well-defined contractual arrangements. Studios traditionally hired professional artists, photographers, and illustrators to create poster art either as employees or as commissioned freelancers. In both cases, copyright ownership was straightforward: when produced by employees, the work qualified as a “work made for hire” under 17 U.S.C. § 101, and when produced by independent contractors, studios routinely obtained ownership through written assignments or licenses. There was little ambiguity about who the author was, nor doubt that a human creator was responsible for the expressive content of the work. Every element could be tied to a known creator or to a properly licensed source, ensuring a clear chain of title. The artists used their own skill and imagination to design the compositions, so there was less risk that components of the poster were inadvertently copied from unlicensed work.

By contrast, AI image generators depend on massive datasets scraped from the internet, which may include millions of copyrighted photographs, illustrations, and prior movie posters. This automated training process blurs the line between inspiration and reproduction and raises novel infringement concerns which did not arise in the analog era.

Performer consent was also built into traditional production practices. Studios already held contractual rights to use actors’ likenesses for marketing, allowing photographers and designers to incorporate approved stills into the final poster. The process was deliberate, and consent driven. With AI, however, a model can produce a highly realistic likeness—or even a “look-alike”—of a performer without authorization, likely violating their right of publicity. This risk is magnified in states, such as California and New York, where statutes like California Civil Code section 3344 specifically protect against unauthorized commercial use of a person’s identity.

Finally, the creative process for traditional posters was transparent and traceable. From concept sketches to photography and lithographic printing, every stage could be documented, and each contributor could be credited. This transparency gave studios legal certainty regarding ownership, attribution, and liability. By contrast, the “black-box” nature of generative AI makes it nearly impossible to identify which training images influenced a given output, complicating questions of provenance and infringement.

Source: Colin Murdoch Studios

Legal Landscape

Under U.S. law, copyright attaches only to works of human authorship. The U.S. Copyright Office reaffirmed this in its 2023 Policy Guidance on Works Containing Material Generated by Artificial Intelligence, stating that “works produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author are not registrable.” Thus, a poster generated entirely by a system, such as Midjourney or DALL·E, would be largely uncopyrightable. This would leave studios without conventional ownership or enforcement rights which were given with earlier artistic works.

However, if a designer uses AI as a tool by providing detailed prompts, selecting from multiple outputs, and compositing elements, the resulting image could have sufficient human authorship to be registered for the human-created aspects only. Courts have begun reinforcing this principle. The District Court in Thaler v. Perlmutter held that an image autonomously created by an AI system could not be copyrighted because “human authorship is a bedrock requirement.” This only further reinforces that the artistic expression of Hollywood through poster art is something worth protecting.

Further, because AI image generators are trained on vast datasets scraped from the internet, rights-holders have already filed suits alleging that training on protected works without consent constitutes infringement. In Getty Images (US), Inc. v. Stability AI, Ltd., Getty alleged that Stability AI copied more than 12 million photographs to train its Stable Diffusion model without authorization. The case, along with a pending U.S. class action by visual artists, may determine whether training-data in this way is fair or infringing. 

Source: "AI-created images lose US copyrights in test for new technology", Business World Online. 

In the United States, the Visual Artists Rights Act of 1990 (VARA) affords some protections for certain visual works. Although VARA applies mainly to unique works of “recognized stature,” the policy implications for this use- case are clear: if an AI-generated poster copies the recognizable style of a living artist, it may undermine that artist’s market and reputation. Even more than this, the Supreme Court’s decision in Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith narrowed the scope of transformative fair use further, warning that a use with “a fundamentally different aesthetic” may still infringe when it serves the same commercial purpose. For movie studios, this means AI outputs which resemble existing posters could expose them to derivative-work or infringement claims,  especially if the new design mimics the composition or expressive style of a previously protected work. The high profitability of blockbuster movies makes movie studios extremely risk averse when it comes to infringement claims of this kind. It is interesting that this has not deterred them from attempting to use this kind of generated artwork without regulation. 

Additionally, AI’s capacity to generate realistic likenesses introduce a variety of publicity concerns. The right of publicity protects individuals against unauthorized commercial exploitation of their name, image, or likeness. If AI tools are used to generate photorealistic depictions of performers without consent, studios risk significant liability under state laws such as California’s Civil Code § 3344.1. Tennessee has even proposed the ELVIS Act which would extend these protections to AI-generated “digital replicas.”

Legal Analysis and Business Implications

Considering these factors, contract law is a seemingly viable safeguard against AI concerns. Studios must require that agencies and vendors disclose any use of AI in their contractual relationships, warrant that outputs do not infringe third-party rights, grant clear and transferable licenses for commercial use, and indemnify the studio against infringement or publicity claims. The creative potential of AI in film marketing is vast, but its legal boundaries are far from settled. To navigate this uncertainty, studios should document human contribution, secure likeness rights, verify company data provenance, and craft detailed clauses in all creative contracts. Until courts or legislatures modernize IP law for machine-assisted creativity, prudent legal counsel remains the most essential design element in every AI-generated movie poster.

  The allure of AI-generated movie posters lies in their speed, novelty, and apparent creativity, however,  true artistry has never been measured by efficiency. Film marketing is undeniably inseparable from the vision of human illustrators, photographers, and designers who transformed simple advertisements into cultural artifacts. From Indiana Jones to Star Wars to Back to the Future, their work has captured not only the image of a film but its emotional resonance. This is something no algorithm can replicate. While AI may offer a new set of tools, it also threatens to dilute the integrity of artistic authorship. It obscures creative provenance and undermines the livelihoods of those whose imagination has given cinema its magic. As studios and lawmakers navigate this technological shift, the goal should not replace artists with algorithms, but to ensure that those tools serve the human creative spirit that has adorned movie posters since their inception.

*The views expressed in this article do not represent the views of Santa Clara University.