YouTube Will Let You Build Your Own Algorithm With AI

YouTube is introducing a new feature called custom feeds that lets users create personalised video recommendations by describing the type of content they want in their own words.

The feature uses Google’s Gemini AI model to build a feed based on user prompts. Instead of relying only on YouTube’s existing recommendation system, users can explain the topics, style, mood, or types of videos they want to see.

Users Can Create Personalised Video Algorithms

To create a custom feed, users enter a description into a prompt box on YouTube. The request can include details about what content to prioritise, what to avoid, and how the feed should be organised.

For example, users could create a feed focused on video podcasts for a daily commute or request relaxing commentary videos for a specific mood. The generated feed will appear as a separate tab at the top of the YouTube home page.

AI-Powered Feed Creation Expands Across Platforms

YouTube’s custom feeds follow similar efforts on other social platforms that let users personalise algorithms through direct instructions. Bluesky introduced custom feeds to let users choose more personalised content experiences.

Other platforms, including Instagram, Threads, X, and Spotify, have also explored tools that give users more control over what appears in their feeds.

YouTube Expands AI Features

YouTube says custom feeds are designed to help users explore the platform’s large video library more easily. The company said its platform contains more than 20 billion videos, creating opportunities for users to discover content through more specific requests.

YouTube introduced the custom feeds feature during its Made On YouTube event. The company said the feature will begin rolling out on web and mobile starting next month.

Custom feeds will not replace YouTube’s main recommendation feed. Instead, users can switch between their regular home feed and personalised, AI-created feeds based on their interests.