Skip to main content

ITmatterss

Facebook’s new AI tool lets users control what appears in Feed and Reels

Vertical Share Bar
your_algorithm

News in short

  • Facebook is rolling out a new AI-powered feature called Your Algorithm.
  • The tool lets users manage topics they want to see more or less in Feed and Reels.
  • Users can describe their content preferences in their own words.
  • New Interested and Not interested options offer more direct feedback on posts.
  • Meta says large language models help interpret users’ interests and update recommendations in real time.

Facebook has introduced Your Algorithm, an AI-powered feature that gives users more control over content recommendations in Feed and Reels. The tool lets people specify topics they want to see more often or reduce in their recommendations.

Meta announced the feature on October 14, 2026. It adds a dedicated customisation section to Facebook’s three-dot menu, allowing users to manage their content preferences directly.

The feature uses large language models (LLMs) to understand preferences expressed in natural language. Instead of relying only on likes, comments and viewing activity, Facebook can also use users’ direct feedback about the content they want to discover.

Meta says the goal is to make recommendations more relevant to individual interests. However, the feature changes recommendations rather than giving users complete control over everything that appears in their feed.

How Facebook’s Your Algorithm feature works

Your Algorithm allows users to add and manage topics that influence recommendations across Feed and Reels. Users can explain what interests them in their own words, giving Facebook additional context about their preferences.

The tool uses LLMs to interpret the interests, context and intent behind these descriptions. This could help the recommendation system distinguish between related topics and adjust what it shows accordingly.

For example, someone interested in smartphone launches but less interested in celebrity content could use their preferences to guide recommendations. The actual results will depend on how Facebook interprets those instructions and the content available on the platform.

Users can access the customisation tool through the three-dot menu on Facebook posts. Meta says recommendations can update in real time as users provide new preferences.

Facebook expands Interested and Not interested feedback

Meta is also expanding feedback options for posts appearing in Feed and Reels. Users can select Interested or Not interested to tell Facebook whether they want to see more or less of a particular type of content.

The platform may ask follow-up questions to understand the reasons behind that feedback. These responses can provide additional context for future recommendations.

This gives users another way to influence their feeds without opening the dedicated Your Algorithm section. It also allows Facebook to collect more specific feedback than passive signals such as viewing time or engagement alone.

The changes could be useful for people whose recommendations no longer match their interests. However, they do not mean every unwanted post will disappear or every preferred topic will appear more frequently.

What the update means for Facebook users

Your Algorithm reflects a shift towards giving users a more direct role in personalising social media feeds. Traditional recommendation systems infer interests from behaviour, including likes, comments and viewing history. Facebook’s new tool adds explicit instructions to that process.

Natural-language controls may make personalisation easier for people who do not want to manage individual settings. The expanded feedback options also provide a quicker way to respond to unwanted recommendations.

The key question is how consistently these preferences affect the content users actually see. Meta says the system can update recommendations in real time, but the company has not provided a quantified measure of how much the feature improves recommendation accuracy.

Facebook is therefore giving users more ways to influence its algorithm, rather than replacing algorithmic recommendations with a fully user-controlled feed.

24

Leave a Reply

Your email address will not be published. Required fields are marked *

logo

Get the latest news instantly

You can change your preferences anytime.