YouTube lets users build personalized feeds using Gemini prompts

By Splitfeed EditorialLast updated Sep 26, 2026, 19:17 UTC

On September 23, 2026, YouTube announced a feature that lets users type a plain-language description of the content they want to see, with Google's Gemini AI then generating a personalized video feed from that input. The update was part of a broader platform refresh that also introduced Gemini-powered video editing tools and creator likeness protection features. The custom feed option represents a deliberate handover of algorithmic curation from YouTube's automated recommendation system to the viewer's own expressed preferences.

The significance of this shift lies in how recommendation algorithms have shaped online behavior for years, often in ways users could not easily inspect or override. YouTube's existing system infers preferences from watch history, likes, and engagement signals, a process that has drawn criticism for promoting addictive content loops and limiting exposure to niche or slower-growing creators. Allowing users to direct the algorithm through text prompts introduces a layer of transparency and intention that was largely absent before. What remains genuinely in dispute is whether Gemini's interpretation of a user's prompt will accurately reflect what that user actually wants, and whether this kind of user-directed curation will reduce or simply relocate the filter-bubble problem.

No reactions from the Pro-AI, Anti-AI, or Middle Ground camps have been published yet on this specific story. Pro-AI voices would typically welcome this as an example of AI expanding individual autonomy and reducing the opacity of platform recommendation systems. Anti-AI commentators would likely raise concerns about data collection tied to user prompts, the commercial incentives Gemini may still embed in curated results, and the broader entrenchment of Google's AI infrastructure in daily media consumption. A middle-ground perspective would probably acknowledge the user-control argument while pressing for clarity on how Gemini weighs prompt instructions against YouTube's advertising and engagement priorities.

The key questions to watch going forward are how YouTube discloses the way Gemini processes and stores user prompts, whether creators report measurable changes in traffic patterns as the feature rolls out, and whether Google publishes any documentation on how commercial considerations interact with prompt-generated feeds.