TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Meta announced Facebook tools that let users describe, in their own words, what they want to see more or less of in Feed and Reels. The company says recommendations can respond in real time, but it has not specified rollout coverage, availability by region or how much influence the controls have over ranking.
Meta announced new Facebook controls that let people describe in their own words what they want to see more or less of in Feed and Reels. The company says its AI-powered recommendation tools can use that feedback to adjust suggestions in real time, giving users a more direct way to influence content selected by Facebook’s ranking systems.
The new feature, called Your algorithm, is a dedicated place to add and manage topics a user wants to see in Feed and Reels. Meta says it can be opened from the three-dot menu, which the company has updated to make recommendation controls easier to reach. The announcement describes the tools as a way to fine-tune recommendations, not as a replacement for Facebook’s existing ranking systems.
Users can also respond to individual posts with “Interested” or “Not interested” and may receive follow-up questions asking for more detail. Meta says people can provide that detail in natural language, rather than relying only on preset feedback options. It says recommendations can then change within the same session, although the announcement does not give examples of response times or explain how long a change remains in effect.
Meta says its ranking systems already use many signals, including likes, comments and viewing behavior, to infer what a person may want to see. The company says large language models can help interpret the interests, context and intent expressed in written feedback and combine that information with existing signals. These descriptions come from Meta’s announcement; the company did not provide independent testing or performance measurements with it.
How User Feedback Could Shape Recommendations
The announcement matters because it offers a more explicit alternative to relying solely on inferred interests. A user can state a preference directly, such as wanting more of a topic or less of another, instead of hoping that likes, viewing patterns or other interactions communicate the same thing. That may be useful when a person’s interests change or when their past activity does not reflect what they want to see now.
For Meta, the tools also point to a change in how people can interact with recommendation systems: written instructions become another input alongside behavioral signals. If the controls work as described, users could have more immediate influence over parts of their Feed and Reels experience. The announcement does not establish how much weight such instructions receive, whether they override other signals, or whether the changes affect recommendations outside those surfaces.
The feature does not mean users choose every post they see. Meta describes the system as continuing to rank and recommend content, with written feedback incorporated into that process. That distinction matters: the tools may make preferences easier to express, but the company has not said they provide complete control over the feed or guarantee that particular content will appear or disappear.
As an affiliate, we earn on qualifying purchases.
From Behavioral Signals to Written Preferences
Facebook recommendations have traditionally been shaped in part by what people do on the service. Meta says its systems consider likes, comments and video viewing, among thousands of other signals, to predict which posts may interest each user. Those signals can suggest a preference, but they do not always reveal why someone engaged with a post or whether they want similar content in the future.
The announced tools add two related routes for feedback: users can manage topics in Your algorithm, or react to a particular recommendation and explain their response. Meta presents large language models as the technology that can interpret free-form wording and its context. The company calls this an evolution of existing ranking, combining new text input with information its systems already use; it has not published technical details about the models or the ranking changes.
Meta’s report is a company announcement, not an independent evaluation of the feature. It describes intended functionality and expected responsiveness, but offers no user study, benchmark or comparison showing how recommendation quality changes after people use the controls.
““You can now fine-tune your experience and interact directly with the algorithms that rank and recommend what you see on Facebook.””
— Meta
social media content recommendation device
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Rollout Details and Control Limits
Meta’s announcement does not say when the rollout will be complete, which countries or account types will receive the tools first, or whether they will be available to all Facebook users. It also does not detail how the topic controls are presented, what kinds of requests the system can interpret, or whether users will be shown a record of the preferences they have entered.
The phrase “real time” is not accompanied by a technical definition or a stated time window. It is also unclear how strongly a written instruction affects ranking compared with a person’s past activity and other signals, how long an adjustment lasts, and whether the same preference applies across Feed and Reels in the same way. Meta has not supplied evidence independent of its own description that would measure how accurately the system interprets user intent.
The announcement likewise does not explain whether written feedback is used only to personalize recommendations or also for other purposes, nor does it describe any privacy or data-retention changes related to the feature. Those questions cannot be answered from the published material.
AI-powered social media feed manager
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Watch for Availability and Usage Details
Facebook users can look for Your algorithm in the three-dot menu as Meta rolls out the feature. The company has not published a detailed launch schedule, so the announcement does not establish when a particular user should expect to see it. Meta may provide additional instructions as access expands, including how to manage topics and send written feedback.
Further details that would clarify the practical effects include the rollout’s geographic reach, whether the controls persist over time, and how the system handles conflicting preferences. Independent reporting or testing after launch could also assess whether recommendations change as Meta describes and whether users find the new controls effective. Until then, the confirmed development is the introduction of these tools and Meta’s stated account of how they are intended to work.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is Facebook’s Your algorithm feature?
Your algorithm is a topic-management area that Meta says lets people add and manage the subjects they want to see in Feed and Reels. It is accessed through the three-dot menu, according to the announcement.
Can users explain what they want in their own words?
Yes. Meta says users can give written feedback after selecting “Interested” or “Not interested”, and may be asked follow-up questions. The company says this feedback can guide recommendations.
Will Facebook recommendations change immediately?
Meta says recommendations respond in real time and can adapt within the same session. The announcement does not define the precise response time or independently verify how quickly changes occur.
Does the feature let users control every post in Feed or Reels?
No such guarantee appears in Meta’s announcement. The company describes feedback as an input to ranking and recommendation systems, not as a way to select every post or ensure that specific content appears or is excluded.
When will the tools be available?
Meta announced the tools in October 2026, but did not provide a detailed schedule, regional availability list or completion date for the rollout.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
