Hollywood & AI have always had a love affair.
The power couple’s first date was almost a hundred years ago, on the set of Metropolis, a dark silent film. In the year 2019, AI (Artificial Intelligence) has become part of everyday life. Algorithms have been used to predict consumer spending habits online for years. And now, Hollywood is looking to harness the power of AI to ensure financial success on high budget films.
There are hundreds of factors that go into making a Hollywood blockbuster. Cast, budget, release date are all critical decisions when developing a movie. The number of datasets that can pulled from a single movie is astounding. Until recently, it was nearly impossible to track, analyze, and provide actionable feedback base on the vast amount of theatrical data.
The newest idea in Hollywood is using AI to determine if a script or particular cast will be profitable. It’s based on machine learning/historical data, and several startup companies are trying to get Hollywood on board with the technology.
Newest Hollywood Hopeful?
It’s an artificial intelligence startup called Cinelytic. The company is one of several startups vying to get the attention of film executives. Should the next James Bond be Idris Ebla or Tom Hiddleston? Cinelytic claims to have the answer and the data to prove it.
The concept is simple but brilliant. Use predictive analytics to cast the actor with the highest ROI or estimate the success of a screenplay prior to production. Calculating ROI for a film used to be a game of chance more suited for Las Vegas than Hollywood. AI evens the odds.
Cinelytic hopes to help avoid large scale blockbuster disasters, such as Waterworld. The 1995 Kevin Costner flop was the most expensive movie ever made. Would switching Costner for Tom Cruise created a box-office success? Cinelytic believes they know the answer.
With movie theater attendance dwindling, and home streaming is increasing, the industry is looking at ways to guarantee profitability. AI may be the answer.
The Final Frontier
What may have seemed like a sci-fi plot just a decade ago is now mainstream. Cinelytic uses predictive analytics as the primary method to determine a film’s earning potential.
In an interview with ITPro, Rob Feng, producer of the 2019 Sundance film, Luce, talks about how the program allowed him to create a stronger movie with a smaller budget.
“It allowed me to try many different scenarios, where normally you’re just stuck with one or two possible approaches, and you’re reliant on a lot of people to go through it. Which is very expensive and time-consuming.”
How does a company like Cinelytic acquire the knowledge and experience to implement this type of advanced AI? The founders, former movie financier Tobias Queisser and Nasa engineer, Dev Sen combined their unique skills.
Sen, who spent the majority of his career at Nasa developing risk models for launches, got the idea on the job. “While there doesn’t appear to be many similarities between assessing the risk of failure of launch vehicles and spacecraft, and the risk of failure of a motion picture, they are quite analogous; they both work with massive budgets, generate a lot of public interest, and occasionally have spectacular failures.”
Together Queisser and Sen have built the ultimate Hollywood prediction tool.
But is it Always Right?
The answer is no.
Algorithms work with historical data, creating blind spots when it comes to accounting for fast-changing cultural tastes. Humans are fickle about their entertainment. A-listers can be downgraded to D-listers for the smallest of missteps.
AI is biased. If the data shows that James Bond has always been a man, then the software is unlikely to suggest a woman. The technology is excellent at running scenarios, but there is still a need for human decision making.
Lack of chemistry is another variable. AI is unable to understand the nuanced human interactions that make a film successful.
Bradford Iten, a 10+ year industry screenwriter veteran, who’s credits include CSI: Miami, has been worried about the dangers of AI technology in screenwriting for a while. In an interview with the Disruptor, he was cautious about the idea of relying solely on AI to determine success.
“One concern is that writers will write to please the algorithm and not write to tell their story. Movie making is still an art form; if you look at the history of film and the changes that have occurred, how does an algorithm account for those societal or even predict the changes in film. Could AI have predicted the rise of the counter culture that influenced film in the late 60s and early ’70s? We would have missed movies like Easy Rider, The Graduate, 2001: A Space Odyssey.”
Iten does believe that the technology can be useful in some situations, specifically when it comes to reviewing potential scripts. “I could see AI being used to clear the herd of submissions. I know script readers who have had to read some truly awful scripts. It would be beneficial and cost-effective to eliminate those.”
Sailing Into the Sunset
In less than a century, Hollywood has seen AI transform from myth to reality. It’s a whole new relationship, one founded in data, not feelings.
When asked what one piece of a caution he would give to studios looking to implement this technology, Iten’s response was telling. “I would advise that it be used as a minimal tool and not lean on it to make final decisions.”
Movies are a lucrative business, but they’re also a form of art. AI startups such as Cinelytic offer a powerful tool, but it’s only a tool. There still needs to be human decision-makers that understand the importance of using film to tell stories that would otherwise go unheard.