Facebook is rolling out a way for users to tell its algorithm what they want to see instead of relying only on likes, comments and scrolling habits.
Meta announced Oct. 13 that it’s rolling out “Your Algorithm,” an AI-powered tool that lets Facebook users influence what appears in their Feed and Reels by describing their preferences in natural language. The feature, which first launched on Instagram, gives users a dedicated space to add and manage topics they care about, accessible through the three-dot menu on every post.
The company also expanded its feedback options. When users tap “Interested” or “Not interested” on a post, Facebook may ask follow-up questions to understand why they liked or disliked something. Users can explain in their own words what they want more or less of. Meta says recommendations update in real time, including within the same session.
How the technology works
Facebook’s ranking systems have long relied on signals like likes, comments, and watch time to predict user preferences—a familiar part of how social media algorithms shape what people see. Now, large language models are interpreting natural language input to understand context and intent.
“LLMs can incorporate the interests, context, and intent behind natural language, unlocking an algorithm you can actually talk to,” Meta wrote in its announcement. “By bringing freeform text together with everything we already understand about what you enjoy, the algorithms behind your Feed and Reels keep getting more relevant and responsive over time.”
The change adds explicit instructions to the signals Facebook already collects from user activity. It gives people another way to influence recommendations, while Facebook continues to decide how content is ranked.
The update follows Facebook’s AI Mode for search, which Meta announced in June to generate answers using public content, including posts from Groups and Reels. Your Algorithm focuses instead on helping users shape recommendations.
More must-read AI coverage
- SS&C Intralinks DealCentre AI vs. Datasite: Which platform is built for the future of dealmaking?
- SS&C Intralinks FundCentre AI vs. Juniper Square: Which platform better supports modern private markets fund managers?
- Why Data, Not Models, Determines AI Success
- The Rise of the AI-Native Factory: How Physical AI Is Transforming Manufacturing
What users can do — and what to expect
The practical benefit is a more direct way to correct unwanted recommendations. Someone researching a temporary interest, such as a home renovation, may want related posts for a while without continuing to see them after the project ends. Explicit feedback provides a way to communicate that change.
When the feature becomes available on your account, open a post’s three-dot menu and look for Your Algorithm to manage your preferred topics. The separate “Interested” and “Not interested” options may also offer follow-up questions where you can explain your feedback.
As a starting point, name the subject you want more or less of: woodworking tutorials rather than celebrity gossip, for example. The broader principle of writing clear AI instructions is useful here, although Meta has not established which wording works best for this feature.
Treat those preferences as guidance rather than a guaranteed filter. Check whether subsequent recommendations reflect your request and adjust your feedback if needed. The announcement does not specify rollout countries, supported languages, or when every account will receive access, so readers should not assume the controls are already available to them.
Read more: Meta is also bringing conversational AI to content creation, with Instagram’s new Edits assistant analyzing Reel performance and suggesting what creators could make next.