Hollywood Group Proposes AI Framework Distinguishing Human-Controlled Use

A coalition of Hollywood filmmakers, studio executives, and Silicon Valley technologists has released its first public framework for integrating AI into film production, proposing that the industry stop treating all artificial intelligence applications as equivalent.

The group, convened quietly in 2023 by producer Kathleen Kennedy and American Film Institute dean Susan Ruskin, has been meeting privately for more than three years to address what participants describe as a "Tower of Babel" problem: the term "AI" has become so broad it describes everything from standard post-production noise reduction to synthetic performances that replace actors entirely.

Their proposed solution centers on a concept called "Human Precision Generative Workflow" (HPG), which distinguishes AI tools that remain under granular human creative control from fully machine-generated content. The framework was detailed in a document released following a weekend gathering of approximately 120 industry participants at AFI.

Why it matters

Hollywood's AI debate has largely stalled in binary terms—either embrace the technology or reject it entirely. This framework attempts to create middle ground by establishing vocabulary that could inform contract negotiations, guild discussions, and audience disclosures. If adopted, it could help the industry move past paralysis toward practical standards for how generative technology enters production workflows without replacing human authorship.

Three categories of AI use

The framework proposes three distinct classifications for AI in final production footage. The first category, "Utility Techniques & Embedded AI," encompasses machine learning tools already present in editing software like Premiere and DaVinci Resolve—functions such as denoising, upscaling, audio separation, and rotoscoping that address technical questions with defined correct answers rather than creative interpretation.

The document deliberately limits its scope to developing shared language around production workflows, acknowledging that equally critical issues around training data, licensing, labor transition, and consent require separate legal and commercial solutions.

From suspicion to common ground

The meetings brought together groups that often regarded one another with suspicion, including Oscar-winning filmmakers, celebrated actors, studio C-suite executives, and major Silicon Valley figures. The first gathering occurred in November 2023 at artist Refik Anadol's Los Angeles studio, in the shadow of Hollywood's contentious labor strikes.

Jessica Sittig, a creative technologist who helped organize the sessions, characterized them as conversations "between these two neighboring tribes who clearly shared a future, but knew very little about each other."

Filmmaker Bryn Mooser, whose production company has explored AI possibilities, framed the sessions as educational: "For filmmakers, it's optional whether you use AI, but it's not optional whether or not to understand it."

The reluctance of many participants to be publicly identified reflects how charged the subject remains in Hollywood. Jon Avnet, who co-chairs the Directors Guild of America's AI Committee with Christopher Nolan, emphasized the importance of proactive engagement: "The key thing was to start the conversation in light of day, and figure out what is happening, and what's going to happen."

The details were first reported by Richard Rushfield in The Ankler, who attended the private meetings alongside editor-in-chief Janice Min under off-the-record terms.