How AI is making movie and TV recommendations better

It’s one of the most common low-risk annoyances of modern life. At the end of the day, you collapse on the couch, finally have a few minutes to watch one of the dozens of amazing shows or movies available to you thanks to the rise of TV and streaming, and you start scrolling. You spend endless evenings opening apps, aimlessly scrolling through endless rows of identical tiles, without actually watching anything. Eventually, you give up and watch. office again.

In this episode The VergecastWe look at why TV and movie recommendations are so complicated, and how AI can make them better. If Spotify can create an endless playlist of music you’ll love, and YouTube and TikTok always have the perfect thing ready, why can’t Netflix, Hulu, and Max get it right?

It turns out that AI can help, at least a little bit. That’s because models like OpenAI, Google, and others have collected a lot of information about movies and shows. Not just titles and genres, but synopses, reviews, summaries, and more from all over the web, so they can synthesize that information and find connections between titles that were previously hard to find. And as the context window grows, these models can actually collect and understand entire movies at once, opening up a whole new way to understand movies.

But ultimately, recommendations are a human problem, because we’re all human. What you want to see and what you like are far more complex and varied than even the best models can understand. As a result, the idea of ​​sitting down and opening Netflix and having exactly the right title appear right away is not going to work out anytime soon. So instead of hoping for the best, let’s look into using AI tools right now to get to your content at least a little bit faster. Watching movies is great, but scrolling through too many movies is seriously overrated.

If you’d like to learn more about everything discussed in this episode, here are some links to get you started.