Did That Viral Tom Cruise, Brad Pitt AI Fight Scene Get A Green-Screen Assist?

A viral AI video featuring a fight scene between Tom Cruise and Brad Pitt made the rounds on the internet this week, with many marveling at how realistic it looked, but at least one digital creative is questioning how much of it was truly spun up by AI.

The fight scene was made by ByteDance’s AI video generator, Seedance 2.0, using a “2 line prompt,” according to filmmaker Ruairi Robinson.

This Tweet is currently unavailable. It might be loading or has been removed.

However, visual effects artist and app developer Aron Peterson was skeptical, “because we can see the camera movement [in the Pitt-Cruise video] and AI video generators are really bad at simulating realistic camera moves, especially handheld shaky cam.”

He acknowledged that the “Seedance 2.0 model is newer and thus more reliable,” but argued that it’s “highly unlikely that just two prompts and thirty seconds were needed to generate a full multi-angle fight scene.”

On Seedance’s website, “it only took 10 seconds to find green screen footage of two stuntmen performing the same fight choreography we see in the Cruise vs Pitt scene,” Peterson writes. “Seedance had used the green screen footage for a different demo—this time using a prompt for an anime style fight scene.”

Some of the reference video is visible in examples on the Seedance homepage. (Credit: Seedance)

The distinction is important because some argued that AI tools like Seedance could kill modern filmmaking. If a simple, two-line prompt can create an action scene out of thin air, why pay for actors, a choreographer, wardrobe stylist, and a full film crew? The green-screen footage, however, suggests that Seedance took real fight-scene footage, generated the background, and added the faces of Pitt and Cruise.

“It’s important to note that not anyone could shoot the green screen video above,” Peterson says. “Video to video requires excellent input/source material for best results, as Seedance did. Hiring a green screen studio, stuntmen, choreographer, lighting crew and cameraman would cost a couple of grand a day on the low end. Out of that all they got was about 20 seconds of footage. Then there is the cost of generating. We don’t know how many times they had unusable output. It is doubtful that it was a one shot generation per angle.”

Recommended by Our Editors

In his post, Peterson tells Seedance and Robinson that “the ball is in your court,” and asks them to explain whether the clip is “really just a 2 line prompt or…2 lines, green screen video footage, and face references too? The evidence appears to show that stuntmen were filmed from several angles, that a clip had to be generated for every angle, and then finally all clips were stitched together for marketing.”

On X, Robinson hasn’t addressed Peterson directly, but doesn’t seem too concerned about the controversy. “Sorry i typed 2 lines and pressed a button, i now know this was wrong,” he joked last week.