Warner Bros has signed a deal with an artificial intelligence company to help it with movie releases. The studio has confirmed it will be using a'revolutionary new AI-driven project management system', launched last year by Cynelytic, a Los Angeles-based AI and cloud tech company. The platform provides forecasting and financial modelling information, predicting box office revenues of potential movie projects. It also has the potential to assist in working out the value of certain stars, and also in scheduling when a movie should be released. According to Business Wire, 'the platform reduces executives' time spent on low-value, repetitive tasks and instead focuses on generating actionable insights for packaging, green-lighting, marketing and distribution decisions in real time'.

AI is about to play more of a role in the movie-making process. Warner Bros. Pictures has unveiled plans to use Cinelytic's AI project management system to assist in making decisions on movies during the "greenlight process." No, it won't have the final say on whether or not a movie goes forward. Rather, this will help the studio predict a movie's revenue, gauge the value of stars and determine when a title should premiere. Studio execs would ultimately have the final say, but the AI could help determine whether a movie is treated as a summer blockbuster or early-in-the-year filler material.

This analysis explores the temporal sequencing of objects in a movie trailer. Temporal sequencing of objects in a movie trailer (e.g., a long shot of an object vs intermittent short shots) can convey information about the type of movie, plot of the movie, role of the main characters, and the filmmakers cinematographic choices. When combined with historical customer data, sequencing analysis can be used to improve predictions of customer behavior. E.g., a customer buys tickets to a new movie and maybe the customer has seen movies in the past that contained similar sequences. To explore object sequencing in movie trailers, we propose a video convolutional network to capture actions and scenes that are predictive of customers' preferences. The model learns the specific nature of sequences for different types of objects (e.g., cars vs faces), and the role of sequences in predicting customer future behavior. We show how such a temporal-aware model outperforms simple feature pooling methods proposed in our previous works and, importantly, demonstrate the additional model explain-ability allowed by such a model.