From Green Screen to Gray Box: AI Is Stepping Onto the Hollywood Sound Stage

On screen, a woman runs toward a cliff and peers anxiously over the edge, teetering as she looks down at a perilous mountain dotted with boulders, her hair flying in the wind.

She’s not, though. I can see her from where I’m standing in the middle of a Disney sound stage in Los Angeles. She’s standing in a “gray box,” a small, walled-off space on the studio floor, neutral-colored walls speckled with small cross marks for positioning. She’s next to a stack of blue-carpeted boxes that serve as a stand-in for whatever large prop that post-production decides to add in digitally later. 

A camera is pointing at her, following her every move. And an almost instant rendering on a screen has placed her on that cliff, perfectly lit, wind tousling her hair, thanks in part to AI.

More from CNET

Green screens and blue screens have been used for more than half a century in TV and filmmaking, allowing post-production crews to insert visual effects and backgrounds behind the actors. But the bright hue can reflect onto objects and skin, and constant adjustments to the backdrop are needed when the director wants to switch to a new angle of the shot.

Filming in a gray box like the one I’m standing next to offers more flexible shots. And combining that with AI-generated or assisted backgrounds speeds up post-production, according to entertainment studio Promise Advanced Imagination, which showcased its filming product during the Disney Accelerator Demo Day last week.

Setting up shots takes less time in gray boxes than it does with a green screen background, Promise co-founder and chief creative officer Dave Clark told me, giving crews more time to be creative because “they’re not bogged down by having to take 2 to 3 hours to change the setup.”

“You’re able to still capture the right performance and the right blocking … this allows me to just focus on the story, focus on the creative and not worry,” Clark said. “You can get multiple takes because you can get them faster, you don’t have to worry about moving the crew, you don’t have to worry about changing lights, everything stays consistent so I can just focus on the story.”

Crews can save time while filming by using these gray boxes. More time (and cost) is saved during post-production by using AI tools to help fill in the backgrounds and replace lighting and objects. 

While some environments are wholly AI-created, most are made by human 3D renderers and visual effects artists working with AI as one of their tools. Using AI to help fill in backgrounds lowers the cost of production but still ensures it has the size and scale of high-quality productions, Clark says, with production companies needing a lot less capital to produce pilots. 

Clark, who’s also a director, insists that everything still “goes through the hands of human artists,” however.

“Much like Disney or Pixar, the process begins with the script, then it goes into storyboard and then (filming),” Clark said. “I think that traditional mindset and that traditional skill set that you only learn over decades of experience is really what amplifies this technology and lets it work for your creative capabilities.”

As filmmaking and animation studios increasingly explore the use of AI tools in streamlining productions, there have been concerns that human jobs will be removed, including across actors, animators, writers, set designers and more.

David Min, vice president of Disney Innovation, told me this system just fills in the background of movies, which is already done with other technologies, such as green screens. This just makes it easier and faster, since lighting changes can even be done in post-production. It means the people on set can focus just on the performances and direction. There will always be directors, writers and actors, he said.

Disney CEO Josh D’Amaro agrees, saying “technology itself is not the goal.”

“We’re being very aggressive about exploring emerging technologies,” D’Amaro said at the event. “It’s to put it in the hands of our creators … to create these incredible and immersive stories.”

Corinne Reichert/CNET

During the demo day at Walt Disney Studios in Burbank, California, Promise CEO George Strompoulis explained to the media that the company — which bills itself as a generative AI studio developing movies, series and visual effects to “support human artistry” — is made up of four major components: Promise Studios can help produce “studio-quality features at a fraction of the cost and timeline”; then there’s the Muse AI technology platform workflow software; the Curious Refuge AI Filmmaking academy, to serve as “a pipeline of the best artists in the world” and to “upscale” existing film industry workers; and The Generation Company, which is Promise’s AI visual effects business.

Promise has done some testing with the Disney teams, and has a series pilot in production with another studio.

“I think it’s easy for people to have the misconception that there’s an AI button that could just make a movie,” said Rob Bredow, senior vice president of creative innovation at Disney. “That is not the approach that Disney’s taking; it’s not the approach that Promise is taking, which is why this has been a good collaboration.”

I watch the screens as the same woman from the cliff edge now appears to be crouching fearfully next to a blown-out car as explosions rain debris around her. 

She’s still right in front of me, squatting next to those blue boxes in the Disney studio sound stage, camera trained on her. I also watch her pick up a screwdriver from the floor. On screen, it transforms into a mysterious scroll out of a Lara Croft-esque adventure movie.

“The ability to do this at cinematic-quality level is something that we’re at the beginning stages of now,” Clark said. “You not only need great filmmakers; you need creative technologists, you need artists who are masters in the AI tools.”

Senior Editor

Corinne Reichert (she/her) grew up in Sydney, Australia and moved to California in 2019. She holds degrees in law and communications, and currently writes news, analysis and features for CNET across the topics of electric vehicles, broadband networks, mobile devices, big tech, artificial intelligence, home technology and entertainment. In her spare time, she watches soccer games and F1 races, and goes to Disneyland as often as possible. See full bio