en.Wedoany.com Reported - Instagram has announced that users can now more precisely control topic preferences for recommended content in the main feed. The app has extended its algorithm personalization features to the main feed interface, allowing users to choose topics they want to see more or less of in recommended content.

Instagram head Adam Mosseri stated that the purpose of this update is to give users more "control" over their app experience. In a post on Threads, he said that empowering users to shape Instagram into a tool that suits them is in the company's best interest. However, this feature only applies to interest-based topics, such as "rescue dogs" or "parenting humor." When attempting to use this feature to request "posts from people I follow," the system returns an error message saying "no results found."

This limitation poses challenges for creators and businesses on the platform. Mosseri is frequently asked why creators' posts cannot consistently reach all or most of their followers. In his post about the new algorithm controls, he acknowledged some users' frustration. He explained that in the past, users relied on their follow lists to shape their experience, but as recommended content dominates the feed, this tool has gradually become ineffective. He believes the industry consensus is that personal feeds have shifted to Stories and direct messages, making it inevitable to accept content from accounts users don't follow. With only one polished post appearing for every 50 friends in the feed, the content lacks sufficient interest, and algorithmic recommendations fill this gap.

Although Instagram does not seem to plan an overall reduction in recommended content (the app still has a dedicated "Following" feed), the company is exploring ways to better consider user preferences. Mosseri revealed that the new personalization features are powered by large language models, which have helped interpret previously elusive algorithms. In the future, these tools may enable deeper personalization or create fully customized versions of the app. Currently, the company is actively supporting personalization requests tailored to different users, moods or atmospheres, and content types.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com