YouTube will let users build their own algorithm with AI

YouTube is introducing custom feeds that let users describe, in their own words, the kinds of videos they want to see, using Google's Gemini AI model to build a personalized feed around that description, according to TechCrunch.

The feature is designed to give viewers more direct control over how the platform's recommendation system surfaces content for them, moving away from relying solely on watch history and engagement signals to determine what appears in a user's feed.

Recommendation algorithms have long been central to how viewers discover content on YouTube, but they have also drawn criticism over concerns about opaque content curation and the potential to push users toward increasingly narrow viewing patterns. Allowing users to explicitly describe their preferences could offer more transparency into how their feed is shaped.

The tool relies on Gemini, Google's family of AI models, which the company has been integrating across its products, including search, workspace applications and now YouTube's core recommendation experience.

TechCrunch's report did not specify when the custom feed feature would be available to all users or whether it would initially launch as a limited test.