YouTube Launches AI-Powered Custom Feeds So Users Can Shape Their Own Recommendations
YouTube’s Gemini-powered feeds let users define what to include, exclude, or prioritize through natural-language prompts, with results pinned as tabs on the home page. Multiple feeds will roll out on web and mobile starting October 2026, alongside the standard feed.
YouTube announced in September 2026 a new feature called custom feeds, which allows users to describe in their own words the kinds of videos they want recommended to them. The feature uses Google’s Gemini AI model to build a personalized feed based on the user’s written prompt.
To create a custom feed, users type a description into a prompt box on YouTube. The platform suggests examples such as a feed of video podcasts suited to a 30-minute commute, or “relaxing commentary videos to unwind with.” Users can specify what content to include or exclude, what to prioritize, and other preferences in as much detail as they choose. The resulting feed is pinned as its own tab at the top of the YouTube home page.
The feature was introduced publicly at YouTube’s Made On YouTube event on Wednesday, September 23, 2026. Emily Moxley, VP of Product Management for Viewer AI, spoke about the reasoning behind the feature ahead of the event. “There’s over 20 billion videos in the YouTube corpus, so you’re always one search away from a treasure trove of topics,” she said, framing custom feeds as a way for users to better explore the platform’s vast library.
Custom feeds will not replace YouTube’s existing main recommendation feed. Instead, users will be able to switch between their custom feeds and the standard feed when opening the app. Support for creating multiple custom feeds will roll out on both web and mobile starting next month.
The move follows similar feed-building features launched by other platforms, including Bluesky, Meta’s Threads, Instagram, X, and Spotify, several of which are also AI-powered. YouTube’s addition suggests the approach of giving users more direct control over their recommendation experience is becoming more widespread across social and content platforms.