Skip to navigationSkip to content

No, Warner Bros. is not letting AI decide what movies it makes

warner bros. tenet
Warner Bros.
Just breathe, it’ll be okay.
  • Adam Epstein
By Adam Epstein

Entertainment reporter

Published Last updated This article is more than 2 years old.

Yesterday, the Hollywood Reporter reported that Warner Bros. signed a deal with an artificial intelligence platform to “guide decision-making at the greenlight stage.” Numerous other outlets picked up the story, sending Twitter and Reddit users into a tizzy as film fans began panicking that AI—and not humans—will soon decide whether or not certain films get made.

The truth is not quite so dire, but it does highlight the growing relationship between technology and Hollywood.

Cinelytic, an analytics and financial modeling tool, uses machine-learning models to help film executives streamline decisions about how, when, and where to best release prospective films. It does not explicitly tell humans which films to make, nor do humans rely on it to figure out if a movie pitch should move to production. It’s more akin to a project management system that speeds up existing processes, informing the actual decision-makers with data that it would take them much longer to parse themselves.

Warner Bros. is now the biggest name in Hollywood to sign on to use Cinelytic. A handful of other studios and production companies, including Sony Pictures and STX Entertainment, are already using the platform.

“This has absolutely nothing to do with creative decisions,” Candice McDonough, Warner’s head of theatrical communications, told Quartz over the phone. She added the partnership is solely with the studio’s international division, which handles marketing and distribution for films outside North America.

Still, some readers saw the news and immediately declared the death of creativity in cinema. For better or worse, though, major film and TV companies like Warner Bros. have been using data to inform key decisions for a long time. Platforms like Cinelytic just make it easier.

Before a film is officially okayed (or “green lit”) by a distributor, teams of industry professionals run all sorts of projections about how much money it might make at the box office, as well as how and when it should be released to optimize reach and profits. These projections—based on things like a film’s cast, the director’s track record, and the popularity of the source material—are used by Hollywood executives every day.

Of course, relying solely on that data could lead to a boringly homogenous slate of films and force companies to miss out on huge opportunities. As Zack Stentz, a screenwriter who co-wrote the films Thor and X-Men: First Class pointed out on Twitter, the historically successful Marvel Cinematic Universe might not exist if film executives didn’t take a chance on an idea the data suggested was extremely dicey. Sometimes there is no substitute for human intuition.

While platforms like Cinelytic might inevitably underrate films like Iron Man, they can help Hollywood avoid more certain disasters. For instance, as a proof of concept, Cinelytic last year projected the Hellboy reboot would fail at the box office. It did. The platform guessed the film would gross just $23.2 million at the US box office on its $50 million budget; it wound up making even less—only $21.9 million. The executives in charge of Hellboy would have been better off using that $50 million on other projects with more potential.

Cinelytic will also come in handy at film festivals, which are increasingly the sites of intense bidding wars between studios and streaming services over hot projects. Deals can be made—or fall apart—in the matter of hours. The platform can turn what might otherwise be days worth of number-crunching into the click of a button. If the AI-driven tech proves to be a helpful asset for a company as significant as Warner Bros., it could become an industry standard. But that’s not necessarily reason to freak out.

📬 Kick off each morning with coffee and the Daily Brief (BYO coffee).

By providing your email, you agree to the Quartz Privacy Policy.