Tech Policy at Duke University

Predictions for how artificial intelligence (AI) will impact labor markets are dominating economic and policy discussions. AI may expose more jobs to automation, leading to job loss and the creation of more specialized job roles. Employers may increasingly hire for AI-augmented roles, requiring job seekers to have stronger AI skills.

In the film industry, AI’s disruption is likely to be more pervasive. AI has the potential to disrupt jobs across film production—from screenwriters to technical crew to actors. Generative AI was a core issue behind the 2023 Screen Actors Guild-American Federation of Television and Radio Artists (SAG-AFTRA) and Writers Guild of America (WGA) strikes where they negotiated protections against use of written material for training generative AI. SAG-AFTRA reached another tentative deal with the Alliance of Motion Picture and Television Producers in early May. The details of the deal are not yet known.

Deepfake actors remain a concern for SAG-AFTRA

Generative AI remains a source of tension.  In 2025, an AI production company called Particle6 introduced an AI actress named Tilly Norwood. That October, Kevin O’Leary of Shark Tank promoted the use of AI extras, arguing that millions of dollars could be saved. He made these statements while promoting his acting debut in Marty Supreme. For SAG-AFTRA, which represents actors, the emergence of deepfake actors has threatened job security. Leading up to the most recent negotiations, SAG-AFTRA’s chief negotiator, Duncan Crabtree-Ireland, pushed for financial penalties that would eliminate cost savings for the use of generative AI actors, which the union calls synthetic performers. The union is also concerned about the use of its members’ performances for training generative AI tools. Actors have some protections against the use of AI in this regard, such as trademarking their voice and likeness. However, these protections may not be broadly accessible to all actors, especially those in minor roles.

The Academy of Motion Pictures Arts and Sciences, commonly abbreviated as The Academy, recently changed and clarified the rules around the use of AI actors in films eligible for the Oscars. Only acting demonstrably performed by humans will be eligible for the prestigious award. The awards body can also request information about the use of AI and the extent of human authorship for films under its consideration. While this is a meaningful step toward limiting the impact of deepfake actors, the prestige films that are typically represented at the Oscars are not representative of all film productions, as not all are submitted for Oscar consideration. The eligibility rules also implicitly protect actors who may be more prominently featured in a film, but leave the door open for the use of AI for more minor acting roles or for extras.

The economics of using AI actors

Cost savings from AI actors go beyond direct wage and technology expenses. Actors, especially in leading roles, may demand high wages for their work, and bring risks that can affect productions. Human actors may be prone to injury or workplace challenges, such as tense relationships with producers and crew members. These challenges can delay filming, and the actors’ reputations can affect a film’s box office performance even after production ends. For a production studio, AI actors carry none of these risks, making them more attractive.

AI actors, however, bear societal costs. Human-produced art requires effort to create something that resonates with audiences; synthetic media threatens creativity. A recent episode of HBO Max’s acclaimed series Hacks sparked conversation about AI’s trivialization of the creative process. The use of deepfake actors normalizes synthetic media, which can have devastating consequences for democracy. Already, deepfake media is used for misinformation, which can impact elections.

Impacts on film crews

While actors are the most visible and recognizable representatives of films, productions employ hundreds of crew members. These members include those responsible for camera work, sound mixing, production design, costume design, lighting, and even logistics. Production workers are concerned about studios’ incentives to use AI to automate and replace them, rather than encourage its use as a tool. They also believe AI compromises the quality and authenticity of production for efficiency gains. In other words, the inhuman nature of AI diminishes the output and leads to soulless art.

Many of these workers are represented by The International Alliance of Theatrical Stage Employees, Moving Picture Technicians, Artists and Allied Crafts of the United States, Its Territories and Canada, commonly referred to as IATSE. Like the other entertainment guilds, IATSE has expressed concern about the impact of AI on worker displacement. In 2023, IATSE released its Core Principles for Applications of Artificial Intelligence and Machine Learning Technology. The principles highlight the union’s embrace of AI as a tool for use by its members, so long as it does not lead to member job loss.

Despite IATSE’s positions, job loss has hit Hollywood’s behind-the-scenes workers. Since 2022, Los Angeles County has lost more than a quarter of its film and TV jobs. From October 2022 to February 2026, production jobs decreased by over 35%.

Film crews have some of the same intellectual property protections afforded to actors. However, crews are not all impacted or protected in the same way, as not all production work creates novel IP. Artists and creatives in the film production process can have producers copyright their work. While these IP rights may protect production workers from having their work used to train generative AI systems, they do not protect workers from replacement by AI.

Recommendations for protecting film workers

Hollywood’s unions and guilds have been actively working to preserve humanness in film productions. Strengthening these efforts will require partnerships and agreements across the Hollywood ecosystem.

The Academy’s recent rule on AI is a meaningful step toward encouraging human authorship for prestige film production. Wider adoption of such rules, which require demonstrable human authorship and allow for interrogation of the use of AI, would encourage more thoughtful and careful use of AI. Adoptions of similar rules by governing bodies like The Golden Globe Foundation, the British Academy of Film and Television Arts, and the Academy of Television Arts & Sciences (which runs the Primetime Emmys for television) could signal a lack of acceptance for entertainment content created substantially by AI. These rules should explicitly name the allowable uses of AI and the extent to which it can be used as a production tool. AI disclosures may have a similar impact, especially for highlighting human authorship in creative categories like makeup, where digital makeup is sometimes used as an enhancement, and score.

Protecting film crews will require a similar approach to SAG-AFTRA’s efforts to penalize the use of synthetic performers. IATSE will need to ensure that using AI is just as expensive as when it is not used. The union has already meaningfully pursued this approach, members’ use of AI is covered under their contract as union work. Minimum staffing requirements, such as those agreed to in the WGA’s 2023 deal, are the most meaningful mechanism for protecting against job loss. As IATSE heads into its next negotiations in 2027, the union should, at a minimum, demand similar protections for behind-the-scenes workers.

The final blog of the series will dive deeper into AI’s innovations in film technology, public perception of AI in film, and the role of awards bodies.