Meta announced on Tuesday the launch of "Your Algorithm," an AI-powered feature that gives Facebook users greater control over their feed content. The feature, which first launched on Instagram, allows users to influence the algorithm that determines what they see. This makes Facebook the latest social media platform to offer users more algorithmic control.
SynthePulse Insight · AI deep reading
Version 1 · 1 source
Meta announced on Tuesday the launch of the AI-driven "Your Algorithm" on Facebook, letting users describe their preferences in natural language to influence recommendations in Feed and Reels. This is an extension of the feature from Instagram to Facebook, and it also marks an attempt by the platform to hand part of recommendation control back to users against the backdrop of AI-dominated content discovery.
Meta announced on Tuesday the launch of a new AI-driven feature called "Your Algorithm" on Facebook. This is not the feature's first appearance—it had already been introduced on Instagram first. This means Facebook users can now also get an algorithm-control entry point similar to what Instagram users have.
Looking at the timeline, Instagram launched Your Algorithm in June, extending customization options to Feed and Explore, after the feature was initially launched for Reels in December 2025. Threads also introduced its own related feature, called "Dear Algo," which allows users to influence the content recommended to them. The three platforms landed one after another, forming a set of moves by Meta in the direction of algorithm control.
The tool allows Facebook users to describe their preferences in natural language, in their own words, thereby influencing what they see in Feed and Reels. The report says users no longer rely entirely on automated recommendations, but can directly express to the platform which topics and content they want to see more or less often.
According to Meta, users can adjust preferences and interact with the algorithm responsible for ranking and recommending posts, and their recommendations will reflect those changes in real time. Meta also said the platform may ask follow-up questions to better understand what users like or dislike, allowing users to provide more detailed feedback.
A distinction needs to be made: the existence of the feature and its entry point are confirmed facts, while "reflecting changes in real time" and "may ask follow-up questions" are statements from Meta, and the report does not provide independent verification.
Meta also updated Facebook's three-dot menu, so that users can manage content preferences directly on individual posts. By opening that menu, users can enter "Your Algorithm," or choose "Interested" and "Not interested" to indicate whether a piece of content matches their preferences.
Putting the control entry point at the individual-post level means users do not have to leave the browsing context to express preferences. But the report does not explain how these preference descriptions are specifically used by the algorithm, what weight they carry, or whether user preferences will be accurately understood and maintained over time.
According to Meta, the feature is intended to make content recommendations more relevant. The report further points out that the launch of this feature reflects Meta's broader effort to give users more influence over their social media experience as AI increasingly shapes content discovery.
The report also says this move marks a departure from the traditional recommendation systems that have shaped the social media experience for years, bringing the platform closer to streaming services that allow users to personalize recommendations based on interests and preferences. These two points are inferential judgments in the report, rather than official statements from Meta, and should be viewed as analysis rather than fact.
The report does not give a specific launch date for the feature on Facebook, saying only that Meta announced it on Tuesday. The report does not provide any user scale, adoption rate, or performance data, nor does it state whether the feature is being rolled out to all Facebook users or in phases.
The report does not disclose the language range or available regions supported by the feature, does not explain how user preference descriptions are specifically used by the algorithm or what weight they carry, and does not provide any third-party verification or independent test results for the feature.
Therefore, regarding whether "Your Algorithm" actually affects recommendations on Facebook in real time as Meta claims, there is currently a lack of independent verification; whether the feature will be rolled out to all Facebook users and when it will be completed is also not made clear by the source. The feature's actual impact on Facebook's content ecosystem and creator distribution is still unknown.
This report is based on a single media report and does not include the original text of Meta's official announcement or additional independent sources. Therefore, statements involving Meta's intent, real-time effects, and the follow-up question mechanism are all statements from Meta or inferences in the report, rather than independently confirmed facts.
Whether Meta will expand this feature to more platforms or functions is not mentioned by the source. Whether preferences described by users in natural language will be accurately understood and maintained over time is also not explained by the source.
This article is based on a single media report (TechCrunch AI, October 13, 2026) and its evidence pack. The feature launch, platform attribution, and entry-point update are confirmed facts; real-time effect, the follow-up question mechanism, and recommendations becoming more relevant are statements from Meta; departing from traditional recommendation systems and moving closer to streaming services are inferences in the report. The report does not provide user scale, adoption rate, performance data, rollout scope, language regions, or third-party verification, and none of those dimensions have been included.
"Your Algorithm" puts part of algorithm control into users' hands, but the report only confirms the existence of the feature and its entry point, not its actual effect. For readers, the key is not the narrative that "Meta gave users control," but the extent of that control, the weight it carries, and whether it takes effect in a verifiable way—these are all still blank for now.
Primary report
Primary source